# data observability

A data-engineering practice for monitoring and improving the health of data in applications and data pipelines.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## ArrivalBench: Agent-Generated Data Pipelines Are Correct Once and Wrong Under Time

DevFeed: [ArrivalBench: Agent-Generated Data Pipelines Are Correct Once and Wrong Under Time](<https://devfeed.tech/articles/arrivalbench-agent-generated-data-pipelines-are-correct-once-and-wrong-under-time-144954.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2610.02363>)

Author: Pranay Kothari

Published: 2026-11-20T05:00:00Z

Content type: news

Language: en

Sources: [arXiv: Databases](<https://devfeed.tech/sources/arxiv-databases.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [streaming data analytics](<https://devfeed.tech/topics/streaming-data-analytics.md>)

Tags: [adversarial](<https://devfeed.tech/tags/adversarial.md>), [agent](<https://devfeed.tech/tags/agent.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [batch](<https://devfeed.tech/tags/batch.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [crash](<https://devfeed.tech/tags/crash.md>), [cs-db](<https://devfeed.tech/tags/cs-db.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [data](<https://devfeed.tech/tags/data.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [delivery](<https://devfeed.tech/tags/delivery.md>), [failure](<https://devfeed.tech/tags/failure.md>), [idempotency](<https://devfeed.tech/tags/idempotency.md>), [log](<https://devfeed.tech/tags/log.md>), [model](<https://devfeed.tech/tags/model.md>), [models](<https://devfeed.tech/tags/models.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [oracle](<https://devfeed.tech/tags/oracle.md>), [ordering](<https://devfeed.tech/tags/ordering.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [time](<https://devfeed.tech/tags/time.md>)

### AI overview

ArrivalBench tests agent-built data pipelines by replaying late, duplicated, out-of-order, and retried records, then comparing the final state with a batch recomputation of the full log. Unlike one-shot snapshot grading, this exposes silent wrong results separately from crashes. Across 40 tasks and eleven models, one-shot grading certified 86-100% of pipelines, while replay found 7.0-79.2% of those certified pipelines silently wrong. Idempotency hazards failed more often than ordering hazards in every model. In one model, a hazard warning reduced silent failures from 48.2% to 10.5%, but raised crashes from 9.0% to 37.0%; total failure fell only from 51.0% to 44.0%. Independently rerunning all eleven arms shifted rates by at most 5.9 percentage points.

### Source excerpt

arXiv:2610.02363v2 Announce Type: replace Abstract: Benchmarks for agent-generated data work grade a pipeline by running it once against a fixed snapshot. ArrivalBench instead re-executes the pipeline an agent leaves behind under adversarial but replayable delivery schedules (late, duplicated, out-of-order and retried records) and requires its final state to equal a batch recomputation of the complete log. Because the oracle recomputes rather than classifies, a wrong table and a crash are distinct verdicts: a crash is visible to monitoring a team already runs, and a wrong table is not. On 40 tasks we built, our reimplementation of single-execution grading certifies 86-100% of the pipelines eleven models produce; re-executing the same artifacts finds 7.0-79.2% of the certified ones silently wrong. The gap is not produced by the repair loop: within the same model and task, pipelines repaired against the snapshot test fail replay about as often as those that passed it first time. In every model, idempotency hazards fail more often than ordering hazards. Separating a wrong answer from a crash also changes how interventions read: a hazard warning cuts one model's silent failure from 48.2% to 10.5% while raising its crash rate from 9.0% to 37.0%, so all-in failure moves only from 51.0% to 44.0%. All eleven arms were independently re-run, and rates moved by at most 5.9 points.

## What Is LLM Monitoring? What to Track Once Your Model Is Live

DevFeed: [What Is LLM Monitoring? What to Track Once Your Model Is Live](<https://devfeed.tech/articles/what-is-llm-monitoring-what-to-track-once-your-model-is-live-141933.md>)

Original publisher: [Read original article](<https://montecarlo.ai/blog-llm-monitoring>)

Author: Ruslan Sultanov

Published: 2026-11-18T18:07:29Z

Content type: article

Language: en

Sources: [Monte Carlo](<https://devfeed.tech/sources/monte-carlo.md>)

Topics: [LLM observability](<https://devfeed.tech/topics/llm-observability.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [ai observability](<https://devfeed.tech/topics/ai-observability.md>), [Monitoring & Alerting](<https://devfeed.tech/topics/monitoring-alerting.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>)

Tags: [agent-trust](<https://devfeed.tech/tags/agent-trust.md>), [agent-trust-glossary](<https://devfeed.tech/tags/agent-trust-glossary.md>), [glossary](<https://devfeed.tech/tags/glossary.md>), [llm](<https://devfeed.tech/tags/llm.md>), [llm-monitoring](<https://devfeed.tech/tags/llm-monitoring.md>), [llm-monitoring-metrics](<https://devfeed.tech/tags/llm-monitoring-metrics.md>), [llm-monitoring-vs-observability](<https://devfeed.tech/tags/llm-monitoring-vs-observability.md>), [llm-performance-monitoring](<https://devfeed.tech/tags/llm-performance-monitoring.md>), [llm-quality-monitoring](<https://devfeed.tech/tags/llm-quality-monitoring.md>), [model](<https://devfeed.tech/tags/model.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [monitoring-llms-in-production](<https://devfeed.tech/tags/monitoring-llms-in-production.md>), [what-is-llm-monitoring](<https://devfeed.tech/tags/what-is-llm-monitoring.md>)

### AI overview

LLM monitoring catches production failures that uptime, latency, and error dashboards miss: an app can be fast and available while returning incorrect or unsafe answers. It continuously tracks performance (latency, tokens, cost, errors), context (source freshness and retrieval), behavior (tool calls, retries, refusals), and output quality (groundedness, relevance, correctness, safety, and user feedback). A practical setup traces requests across retrieval, model, and tool calls; scores traffic with deterministic checks and LLM-as-judge evaluations; alerts on trends or anomalies; then uses trace context to investigate causes such as stale data, prompt changes, or model updates. Evals assess fixed examples before release, monitoring detects production changes, and observability supplies traces and lineage to explain them. The article recommends versioning traces, establishing baselines, checking judges against human review, and assigning alert owners.

### Source excerpt

Your LLM app has been live for a month. Latency is under a second, the error rate is zero, and uptime is perfect. Meanwhile, a product manager forwards a screenshot: the bot told a customer about a discount that ended in March. Nothing on the dashboard moved. That's the gap LLM monitoring fills. This guide ...

## Composing data pipelines in an AI-native workflow

DevFeed: [Composing data pipelines in an AI-native workflow](<https://devfeed.tech/articles/composing-data-pipelines-in-an-ai-native-workflow-136765.md>)

Original publisher: [Read original article](<https://wf.fivetran.com/blog/composing-data-pipelines-in-an-ai-native-workflow>)

Author: Jeffrey Wong

Published: 2026-11-14T00:25:10Z

Content type: article

Language: en

Sources: [Fivetran Blog](<https://devfeed.tech/sources/fivetran-blog.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [Terminal](<https://devfeed.tech/topics/terminal.md>), [DevAIOps](<https://devfeed.tech/topics/devaiops.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Grafana](<https://devfeed.tech/topics/grafana.md>), [Linear](<https://devfeed.tech/topics/linear.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-native](<https://devfeed.tech/tags/ai-native.md>), [cli](<https://devfeed.tech/tags/cli.md>), [command-line](<https://devfeed.tech/tags/command-line.md>), [composability](<https://devfeed.tech/tags/composability.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [fivetran](<https://devfeed.tech/tags/fivetran.md>), [terminal](<https://devfeed.tech/tags/terminal.md>), [workflow](<https://devfeed.tech/tags/workflow.md>)

### AI overview

The Fivetran CLI lets a coding harness manage data connections and schemas, run syncs, and build custom connectors from the terminal, while dbt handles transformations and a harness coordinates the workflow. MCP servers provide context from systems such as ticket trackers and observability tools; the CLI provides scriptable actions and structured output. The described loop can turn a ticket into reviewed connection and model changes, validate them, open a PR, and respond to sync failures. Connection and schema configurations live in the repository for diff-based review and rollback. The text recommends scoped API keys, keeping confirmation prompts for destructive actions, and promoting changes through CI with human approval. Resource-management commands are in beta.

### Source excerpt

The Fivetran CLI slots into a terminal-native loop with dbt so you can stand up a pipeline, change your mind, and redeploy in minutes.

## RevOps Is Quietly Becoming an Engineering Job

DevFeed: [RevOps Is Quietly Becoming an Engineering Job](<https://devfeed.tech/articles/revops-is-quietly-becoming-an-engineering-job-133416.md>)

Original publisher: [Read original article](<https://hackernoon.com/revops-is-quietly-becoming-an-engineering-job?source=rss>)

Author: Malek Damiri

Published: 2026-11-13T22:51:10Z

Content type: opinion

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>), [Software Engineering](<https://devfeed.tech/topics/software-engineering.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [Job](<https://devfeed.tech/topics/job.md>), [metric-standardization](<https://devfeed.tech/topics/metric-standardization.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [release engineering](<https://devfeed.tech/topics/release-engineering.md>)

Tags: [actions](<https://devfeed.tech/tags/actions.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [airflow](<https://devfeed.tech/tags/airflow.md>), [automation](<https://devfeed.tech/tags/automation.md>), [bigquery](<https://devfeed.tech/tags/bigquery.md>), [billing](<https://devfeed.tech/tags/billing.md>), [business-logic](<https://devfeed.tech/tags/business-logic.md>), [careers](<https://devfeed.tech/tags/careers.md>), [crm](<https://devfeed.tech/tags/crm.md>), [cron](<https://devfeed.tech/tags/cron.md>), [data](<https://devfeed.tech/tags/data.md>), [data-analytics](<https://devfeed.tech/tags/data-analytics.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [future-of-work](<https://devfeed.tech/tags/future-of-work.md>), [learn](<https://devfeed.tech/tags/learn.md>), [numbers](<https://devfeed.tech/tags/numbers.md>), [operations](<https://devfeed.tech/tags/operations.md>), [revenue](<https://devfeed.tech/tags/revenue.md>), [revenue-operations](<https://devfeed.tech/tags/revenue-operations.md>), [revops](<https://devfeed.tech/tags/revops.md>), [sales-operations](<https://devfeed.tech/tags/sales-operations.md>), [staging](<https://devfeed.tech/tags/staging.md>)

### AI overview

RevOps is increasingly responsible for operating revenue-related code as well as managing CRM, reporting, and forecasting: warehouse data calls for SQL, and cheap AI-generated scripts still need reliable scheduling, failure handling, and ownership. The author recommends learning SQL, one scheduler (such as cron, launchd, GitHub Actions, or Airflow), version control for number-producing work, data checks, staging and promotion, agreed written metric definitions, and sandboxed AI-agent use. He draws a boundary: data engineering owns shared platforms and infrastructure; RevOps owns business logic and last-mile reporting and automation. Automations whose failure could harm the company should move to the platform with monitoring, on-call ownership, and a second knowledgeable person, avoiding fragile jobs that only one person understands.

### Source excerpt

Malek Damiri, a head of revenue operations, on why RevOps now means schedulers, staging, tested numbers, and AI agents, and what to learn next.

## Schema Drift vs. Semantic Drift in Data Pipelines

DevFeed: [Schema Drift vs. Semantic Drift in Data Pipelines](<https://devfeed.tech/articles/schema-drift-vs-semantic-drift-in-data-pipelines-133351.md>)

Original publisher: [Read original article](<https://www.freecodecamp.org/news/schema-drift-vs-semantic-drift-in-data-pipelines/>)

Author: Yemi Ojedapo

Published: 2026-11-13T11:23:18Z

Content type: article

Language: en

Sources: [Free Code Camp](<https://devfeed.tech/sources/freecodecamp-programming-tutorials-python-javascript-git-more.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Pydantic](<https://devfeed.tech/topics/pydantic.md>), [schema design](<https://devfeed.tech/topics/schema-design.md>), [Data analysis](<https://devfeed.tech/topics/data-analysis.md>), [reproducibility](<https://devfeed.tech/topics/reproducibility.md>)

Tags: [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [drift](<https://devfeed.tech/tags/drift.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [pydantic](<https://devfeed.tech/tags/pydantic.md>), [python](<https://devfeed.tech/tags/python.md>), [schema](<https://devfeed.tech/tags/schema.md>), [schema-drift](<https://devfeed.tech/tags/schema-drift.md>), [semantic](<https://devfeed.tech/tags/semantic.md>), [semantic-drift](<https://devfeed.tech/tags/semantic-drift.md>)

### AI overview

In a synthetic pipeline lab with nine drift scenarios, schema checks caught all five structural changes, but missed semantic shifts that preserved field names and types. Such shifts let the pipeline run while changing results: timezone loss moved items to the wrong day, nulls replaced with empty strings reduced the unequipped count from 205 to 0, and tenfold durability scaling inflated the average. Across the four detectors, strict Pydantic validation caught the most cases (7/9), followed by schema diff (6), loose validation (4), and statistical profiling (2). Each had blind spots: strict rules need domain constraints, while statistical thresholds can raise false alarms and vary by field. Combining detectors improved coverage, but none establishes meaning without producer-consumer agreement on units, timezone conventions, valid values, and null semantics. The experiment used synthetic independent data and one fixed baseline, so its measured false-alarm rates may not generalize to real pipelines.

### Source excerpt

When a column is removed from a source database or dataset, the next pipeline run usually fails, making the problem easy to detect. But when a field keeps the same name and data type but its meaning c

## Composing data pipelines in an AI-native workflow

DevFeed: [Composing data pipelines in an AI-native workflow](<https://devfeed.tech/articles/composing-data-pipelines-in-an-ai-native-workflow-129442.md>)

Original publisher: [Read original article](<https://www.fivetran.com/blog/composing-data-pipelines-in-an-ai-native-workflow>)

Author: Jeffrey Wong

Published: 2026-11-11T20:38:35Z

Content type: article

Language: en

Sources: [Fivetran Blog](<https://devfeed.tech/sources/fivetran-blog.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [Terminal](<https://devfeed.tech/topics/terminal.md>), [DevAIOps](<https://devfeed.tech/topics/devaiops.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Grafana](<https://devfeed.tech/topics/grafana.md>), [Linear](<https://devfeed.tech/topics/linear.md>)

Tags: [ai-native](<https://devfeed.tech/tags/ai-native.md>), [cli](<https://devfeed.tech/tags/cli.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [fivetran](<https://devfeed.tech/tags/fivetran.md>), [workflow](<https://devfeed.tech/tags/workflow.md>)

### AI overview

The Fivetran CLI provides scriptable terminal commands for managing connections, schemas, syncs, and custom connectors, while dbt handles transformations and a coding harness coordinates both. In the described workflow, MCP servers supply context from tickets and operational systems; the harness reads that context, calls CLI and dbt commands, and returns changes for review as code. The CLI exposes discoverable commands, structured JSON output, and exit status, and can be paired with dbt Wizard or general-purpose harnesses. The article's example moves from a ticket to a Stripe-to-Snowflake revenue model, with iteration on schema and model changes and recovery from sync failures. It recommends scoped API keys, keeping confirmation prompts for destructive actions, and reviewing and promoting configuration and model changes through diffs and CI. Resource-management commands are in beta.

### Source excerpt

The Fivetran CLI slots into a terminal-native loop with dbt so you can stand up a pipeline, change your mind, and redeploy in minutes.

## Building Data Observability for Banking Pipelines

DevFeed: [Building Data Observability for Banking Pipelines](<https://devfeed.tech/articles/building-data-observability-for-banking-pipelines-119757.md>)

Original publisher: [Read original article](<https://hackernoon.com/building-data-observability-for-banking-pipelines?source=rss>)

Author: Sunilkumar Reddy Eraganeni

Published: 2026-11-05T17:39:30Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [data observability](<https://devfeed.tech/topics/data-observability.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [observability](<https://devfeed.tech/topics/observability.md>)

Tags: [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [added](<https://devfeed.tech/tags/added.md>), [alert](<https://devfeed.tech/tags/alert.md>), [alerting](<https://devfeed.tech/tags/alerting.md>), [banking](<https://devfeed.tech/tags/banking.md>), [banks](<https://devfeed.tech/tags/banks.md>), [building](<https://devfeed.tech/tags/building.md>), [can](<https://devfeed.tech/tags/can.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-governance](<https://devfeed.tech/tags/data-governance.md>), [data-lineage](<https://devfeed.tech/tags/data-lineage.md>), [data-observability](<https://devfeed.tech/tags/data-observability.md>), [data-pipeline](<https://devfeed.tech/tags/data-pipeline.md>), [data-quality](<https://devfeed.tech/tags/data-quality.md>), [data-reconciliation](<https://devfeed.tech/tags/data-reconciliation.md>), [green](<https://devfeed.tech/tags/green.md>), [here](<https://devfeed.tech/tags/here.md>), [numbers](<https://devfeed.tech/tags/numbers.md>), [observability](<https://devfeed.tech/tags/observability.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [regulatory](<https://devfeed.tech/tags/regulatory.md>), [regulatory-reporting](<https://devfeed.tech/tags/regulatory-reporting.md>), [reports](<https://devfeed.tech/tags/reports.md>), [ship](<https://devfeed.tech/tags/ship.md>), [show](<https://devfeed.tech/tags/show.md>)

### AI overview

Banking pipelines can report success while delivering stale, incomplete, or skewed data, so observability should check the data at each processing boundary, not just whether jobs ran. The article recommends tracking freshness by business date, calendar-aware volume, schema changes, distributions, and lineage; reconciling balances to source is a practical way to catch dropped records. It suggests starting with one regulatory report, adding freshness, volume, and reconciliation checks, assigning named owners, and running alerts in warning-only mode for two weeks before making critical checks block publication. Blocking regulatory outputs versus publishing with a warning depends on deadline tradeoffs. These checks cannot detect fresh, consistent data that is wrong because business logic misapplies a rule; that requires code review, testing, and domain expertise. The SQL volume-check example is explicitly a sketch that does not account for holidays.

### Source excerpt

Pipelines that show green can still ship wrong numbers. Here is how banks can catch silent data failures before regulatory reports break.

## Building pipelines has become easy. Knowing if they work correctly is now easy, too

DevFeed: [Building pipelines has become easy. Knowing if they work correctly is now easy, too](<https://devfeed.tech/articles/building-pipelines-has-become-easy-knowing-if-they-work-correctly-is-now-easy-too-115755.md>)

Original publisher: [Read original article](<https://dlthub.com/blog/introducing-pipeline-observability>)

Author: Adrian Brudaru

Published: 2026-11-03T11:30:00Z

Content type: release

Language: en

Sources: [DLT Hub](<https://devfeed.tech/sources/dlt-hub.md>)

Topics: [data observability](<https://devfeed.tech/topics/data-observability.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-harness](<https://devfeed.tech/tags/ai-harness.md>), [alert](<https://devfeed.tech/tags/alert.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [api](<https://devfeed.tech/tags/api.md>), [availability](<https://devfeed.tech/tags/availability.md>), [context](<https://devfeed.tech/tags/context.md>), [context-graph](<https://devfeed.tech/tags/context-graph.md>), [data](<https://devfeed.tech/tags/data.md>), [general-availability](<https://devfeed.tech/tags/general-availability.md>), [graph](<https://devfeed.tech/tags/graph.md>), [interface](<https://devfeed.tech/tags/interface.md>), [missing-data](<https://devfeed.tech/tags/missing-data.md>), [observability](<https://devfeed.tech/tags/observability.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [product](<https://devfeed.tech/tags/product.md>), [ui](<https://devfeed.tech/tags/ui.md>), [updates](<https://devfeed.tech/tags/updates.md>)

### AI overview

dltHub's observability UI is generally available, with pipeline and run views plus configurable alerts to help teams investigate incomplete or unexpected data even when a run succeeds. The workspace overview highlights jobs needing attention; pipeline charts track success rate, rows, bytes, and duration over time. Run details show job pipelines and their row counts and statuses, while pipeline-run details list loaded tables, rows, bytes, duration, and logs. Teams can alert on failed runs and optionally successful ones by email or Slack, scoped to all jobs or selected pipelines; each alert links to the run and logs. The article notes that changes in row counts can flag issues such as late source data or repeated records, but require investigation rather than proving a failure. Run history, traces, and logs are also available to engineers through the UI and agents through MCP.

### Source excerpt

We are releasing our new observability UI, the interface to the dltHub context graph, to General Availability. Find out how this will help you solve missing data when everything looks right at the surface from your pipelines.

## Benchmarking Claude Haiku, Sonnet, and Opus on 64 Broken Data Pipelines

DevFeed: [Benchmarking Claude Haiku, Sonnet, and Opus on 64 Broken Data Pipelines](<https://devfeed.tech/articles/benchmarking-claude-haiku-sonnet-and-opus-on-64-broken-data-pipelines-114850.md>)

Original publisher: [Read original article](<https://hackernoon.com/benchmarking-claude-haiku-sonnet-and-opus-on-64-broken-data-pipelines?source=rss>)

Author: Akshay Sharma

Published: 2026-11-03T11:18:57Z

Content type: news

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [Data Quality](<https://devfeed.tech/topics/data-quality.md>), [Claude](<https://devfeed.tech/topics/claude.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent-evaluation](<https://devfeed.tech/tags/ai-agent-evaluation.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-benchmarking](<https://devfeed.tech/tags/ai-benchmarking.md>), [batch](<https://devfeed.tech/tags/batch.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [broken-data-pipelines](<https://devfeed.tech/tags/broken-data-pipelines.md>), [claude](<https://devfeed.tech/tags/claude.md>), [claude-haiku](<https://devfeed.tech/tags/claude-haiku.md>), [claude-opus](<https://devfeed.tech/tags/claude-opus.md>), [claude-sonnet](<https://devfeed.tech/tags/claude-sonnet.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-pipeline-recovery](<https://devfeed.tech/tags/data-pipeline-recovery.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [haiku](<https://devfeed.tech/tags/haiku.md>), [haiku-vs-sonnet-vs-opus](<https://devfeed.tech/tags/haiku-vs-sonnet-vs-opus.md>), [incidents](<https://devfeed.tech/tags/incidents.md>), [opus](<https://devfeed.tech/tags/opus.md>), [repair](<https://devfeed.tech/tags/repair.md>), [safely](<https://devfeed.tech/tags/safely.md>)

### AI overview

In this synthetic benchmark of 64 DuckDB pipeline incidents, Sonnet and Opus tied when change notes were available, while Opus handled all 13 tested cases when notes were removed and Sonnet dropped two; Opus cost about four times as much. Haiku more often escalated fixable incidents and made two fixes that ran but produced incorrect data. All models escalated incidents with no safe fix, but sparse sample rows sometimes hid conflicting duplicate values, leading all three to silently choose a wrong amount. Ten data quality monitors did not catch the benchmark's corrupted fixes. The author recommends starting with Sonnet, using a human-review confidence floor (0.65 yielded zero harmful Sonnet fixes in this dataset while solving 47-49 of 52 fixable incidents), checking results against raw or trusted data, and improving source change notes. These findings come from one synthetic pipeline, one model family, and one run per case; the threshold was tuned on the same data.

### Source excerpt

Can AI agents safely repair broken data pipelines? See how Claude Haiku, Sonnet, and Opus performed across 64 simulated data engineering incidents.

## Data Engineering Weekly #294

DevFeed: [Data Engineering Weekly #294](<https://devfeed.tech/articles/data-engineering-weekly-294-112997.md>)

Original publisher: [Read original article](<https://www.dataengineeringweekly.com/p/data-engineering-weekly-294>)

Author: Ananth Packkildurai

Published: 2026-11-02T05:33:47Z

Content type: news

Language: en

Sources: [Data Engineering Weekly](<https://devfeed.tech/sources/data-engineering-weekly.md>)

Topics: [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Integration Strategies](<https://devfeed.tech/topics/ai-integration-strategies.md>), [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>), [Software Engineering](<https://devfeed.tech/topics/software-engineering.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-framework](<https://devfeed.tech/tags/agent-framework.md>), [agent-state](<https://devfeed.tech/tags/agent-state.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-ml](<https://devfeed.tech/tags/ai-ml.md>), [ai-systems](<https://devfeed.tech/tags/ai-systems.md>), [airflow](<https://devfeed.tech/tags/airflow.md>), [alert](<https://devfeed.tech/tags/alert.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [apache](<https://devfeed.tech/tags/apache.md>), [apache-airflow](<https://devfeed.tech/tags/apache-airflow.md>), [api](<https://devfeed.tech/tags/api.md>), [api-server](<https://devfeed.tech/tags/api-server.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [newsletter](<https://devfeed.tech/tags/newsletter.md>), [weekly](<https://devfeed.tech/tags/weekly.md>)

### AI overview

The excerpt highlights practical approaches to making data platforms and AI systems more reliable. Zepto precomputes shared Superset datasets for dashboard reuse, reporting a 30-35% reduction in warehouse costs without changing analysts' workflows. Razorpay migrated more than 2,500 DAGs to Airflow 3 incrementally, running versions side by side; incidents also led it to separate task traffic from the UI. Meta's PrivacyLib applies reusable monitoring or blocking checks at data access points, including controls for internal AI agents on supported paths. Medium replaced its sensitive-story classifier with an LLM guided by written criteria and human-labeled evaluation examples, while retaining separate models for spam and topics because trend adaptation and processing costs mattered there. For AI agents, the excerpt stresses linking pipeline runs to logs and metadata, recording business context, and limiting permissions. Teradata's Loom adds resumable agent state and controls for tool access, spending, and approvals. A DataFusion lookup workload saw nearly eightfold higher throughput from combined changes, including an experimental planner optimization unavailable in DataFusion at the--

### Source excerpt

The Weekly Data Engineering Newsletter

## Introducing Flight Explainer: Live charts of your python pipelines

DevFeed: [Introducing Flight Explainer: Live charts of your python pipelines](<https://devfeed.tech/articles/introducing-flight-explainer-live-charts-of-your-python-pipelines-113420.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/flight-explainer>)

Author: Hamilton Ulmer; Colin Rogers

Published: 2026-11-02T00:00:00Z

Content type: article

Language: en

Sources: [MotherDuck](<https://devfeed.tech/sources/motherduck.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Data visualization](<https://devfeed.tech/topics/data-visualization.md>), [Python](<https://devfeed.tech/topics/python.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [analyze](<https://devfeed.tech/tags/analyze.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [branch](<https://devfeed.tech/tags/branch.md>), [build](<https://devfeed.tech/tags/build.md>), [can](<https://devfeed.tech/tags/can.md>), [charts](<https://devfeed.tech/tags/charts.md>), [explain](<https://devfeed.tech/tags/explain.md>), [explainer](<https://devfeed.tech/tags/explainer.md>), [flight](<https://devfeed.tech/tags/flight.md>), [know](<https://devfeed.tech/tags/know.md>), [live](<https://devfeed.tech/tags/live.md>), [minutes](<https://devfeed.tech/tags/minutes.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [plan](<https://devfeed.tech/tags/plan.md>), [python](<https://devfeed.tech/tags/python.md>), [run](<https://devfeed.tech/tags/run.md>)

### AI overview

Flight Explainer adds live run charts to MotherDuck Flights, showing a declared workflow alongside what happened at each step. The chart marks running, completed, skipped, and failed steps; traces the branch taken; and separates whether a step succeeded from the decision it made. Steps can report live and retained progress as fractions, percentages, or per-item fan-out results, including durations and errors. They also record tables, files, and endpoints touched. The tool compares declared work types, such as transform, with reported effects and flags mismatches as drift. The source says no installation or configuration is needed: ask an agent to instrument a flight and display a run in chat.

### Source excerpt

Agents can build a MotherDuck Flight in minutes. But how do you know it does what you asked? Flight Explainer is EXPLAIN ANALYZE for your pipelines: a live chart of every run showing the plan, which branch it took, how far each step got, and what it touched.

## Your Agents Will Ship Before Your Homegrown Monitoring Does

DevFeed: [Your Agents Will Ship Before Your Homegrown Monitoring Does](<https://devfeed.tech/articles/your-agents-will-ship-before-your-homegrown-monitoring-does-109692.md>)

Original publisher: [Read original article](<https://montecarlo.ai/blog-homegrown-monitoring>)

Author: Ruslan Sultanov

Published: 2026-10-30T16:35:13Z

Content type: opinion

Language: en

Sources: [Monte Carlo](<https://devfeed.tech/sources/monte-carlo.md>)

Topics: [ai observability](<https://devfeed.tech/topics/ai-observability.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [cloud cost management](<https://devfeed.tech/topics/cloud-cost-management.md>)

Tags: [agent-trust](<https://devfeed.tech/tags/agent-trust.md>), [agent-trust-ai-observability](<https://devfeed.tech/tags/agent-trust-ai-observability.md>), [agentic-operations](<https://devfeed.tech/tags/agentic-operations.md>), [agents](<https://devfeed.tech/tags/agents.md>), [agents-in-production](<https://devfeed.tech/tags/agents-in-production.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-coding-tools](<https://devfeed.tech/tags/ai-coding-tools.md>), [ai-observability](<https://devfeed.tech/tags/ai-observability.md>), [budget](<https://devfeed.tech/tags/budget.md>), [build](<https://devfeed.tech/tags/build.md>), [build-vs-buy](<https://devfeed.tech/tags/build-vs-buy.md>), [building](<https://devfeed.tech/tags/building.md>), [challenges](<https://devfeed.tech/tags/challenges.md>), [code](<https://devfeed.tech/tags/code.md>), [coding](<https://devfeed.tech/tags/coding.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [cto](<https://devfeed.tech/tags/cto.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [homegrown-monitoring](<https://devfeed.tech/tags/homegrown-monitoring.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [production](<https://devfeed.tech/tags/production.md>), [ship](<https://devfeed.tech/tags/ship.md>)

### AI overview

The article argues that homegrown agent monitoring can suit teams with one or two agents, but often falls behind as deployments grow. Around a fifth agent, it says, token costs, tooling maintenance, incident investigation gaps, and higher-impact failures tend to converge. A trust platform involves monitoring data quality and lineage, agent context, tool use, cost and answer quality, plus agentic operations to profile data, set checks, triage alerts and fix issues. Coding tools may speed initial implementation, but tuning baselines and reducing alert noise depend on accumulated response history. In the authors' mid-sized-enterprise model, matching a bought platform takes about 24 months, with agentic operations beginning in month 19; a three-year build for a composite company with 30,000 tables is estimated at $5.14M versus $0.98M to buy, peaking at 7-10 engineers. The article advises estimating agent growth, assigning ongoing maintenance ownership, modeling evaluation costs, and deciding how to manage coverage before agentic operations are available. Estimates vary by organization.

### Source excerpt

Your first agent went to production in a few weeks. The plan for monitoring it was reasonable: build it ourselves. You know your stack, you can start today, and there's no procurement cycle in the way. Surely, homegrown monitoring makes sense, right? Then the second agent shipped, and the fifth. The monitoring is still on ...

## Article: InfoQ AI, ML and Data Engineering Trends Report - 2026

DevFeed: [Article: InfoQ AI, ML and Data Engineering Trends Report - 2026](<https://devfeed.tech/articles/article-infoq-ai-ml-and-data-engineering-trends-report-2026-107540.md>)

Original publisher: [Read original article](<https://www.infoq.com/articles/ai-ml-data-engineering-trends-2026/>)

Author: Srini Penchikala, Chip Huyen, Daniel Dominguez, Hien Luu, Arthur Casals

Published: 2026-10-30T09:00:00Z

Content type: article

Language: en

Sources: [InfoQ](<https://devfeed.tech/sources/infoq.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [AI Engineering](<https://devfeed.tech/topics/ai-engineering.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Humanoid Robots](<https://devfeed.tech/topics/humanoid-robots.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [Test-driven development](<https://devfeed.tech/topics/tdd.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [adoption](<https://devfeed.tech/tags/adoption.md>), [agent2agent](<https://devfeed.tech/tags/agent2agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-and-ml](<https://devfeed.tech/tags/ai-and-ml.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [ai-development](<https://devfeed.tech/tags/ai-development.md>), [ai-ml](<https://devfeed.tech/tags/ai-ml.md>), [ai-ml-data-engineering](<https://devfeed.tech/tags/ai-ml-data-engineering.md>), [ai-ml-data-engineering-trends-2026](<https://devfeed.tech/tags/ai-ml-data-engineering-trends-2026.md>), [article](<https://devfeed.tech/tags/article.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [code-generation](<https://devfeed.tech/tags/code-generation.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [infoq](<https://devfeed.tech/tags/infoq.md>), [infoq-trends-report](<https://devfeed.tech/tags/infoq-trends-report.md>), [infoq-trends-report-2026](<https://devfeed.tech/tags/infoq-trends-report-2026.md>), [large-language-models](<https://devfeed.tech/tags/large-language-models.md>), [ml](<https://devfeed.tech/tags/ml.md>), [ml-data-engineering](<https://devfeed.tech/tags/ml-data-engineering.md>), [model-context-protocol-mcp](<https://devfeed.tech/tags/model-context-protocol-mcp.md>), [report](<https://devfeed.tech/tags/report.md>), [robotics](<https://devfeed.tech/tags/robotics.md>), [trends](<https://devfeed.tech/tags/trends.md>)

### AI overview

The excerpt shows AI adoption shifting from model development toward operating agents and inference systems in production. Agents are expanding from personal assistants to team and organization wide systems, increasing the need for communication protocols, runtime and harness capabilities, and evaluations that assess the whole workflow--including task completion, cost, latency, interventions, and safety. Inference engines such as vLLM and SGLang are making it more approachable to host open weight models internally, including for sensitive data or cost control. The report moves small and reasoning models, agentic coding, MCP, and inference optimization into Early Adopters; generative AI, RAG, vector databases, and context engineering into Early Majority; and several mature technologies, including Spark and lakehouses, into Late Majority. It also highlights physical AI, where reliability, simulation, safety, and evaluation matter because errors can have physical consequences. These findings reflect the supplied excerpt; it does not provide the report's full detail.

### Source excerpt

This InfoQ Trends Report offers readers a comprehensive overview of emerging trends and technologies in the areas of AI, ML, and Data Engineering. This report summarizes the InfoQ editorial team's and external guests' view on the current trends in AI and ML technologies and what to look out for in the next 12 months. By Srini Penchikala, Chip Huyen, Daniel Dominguez, Hien Luu, Arthur Casals

## Monte Carlo Data launches Bayes, an AI agent for autonomously monitoring and fixing autonomous agents

DevFeed: [Monte Carlo Data launches Bayes, an AI agent for autonomously monitoring and fixing autonomous agents](<https://devfeed.tech/articles/monte-carlo-data-launches-bayes-an-ai-agent-for-autonomously-monitoring-and-fixing-autonomous-agents-104540.md>)

Original publisher: [Read original article](<https://siliconangle.com/2026/10/28/monte-carlo-data-launches-bayes-an-ai-agent-for-autonomously-monitoring-and-fixing-autonomous-agents/>)

Author: Mike Wheatley

Published: 2026-10-28T13:00:50Z

Content type: news

Language: en

Sources: [SiliconANGLE](<https://devfeed.tech/sources/siliconangle.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [ai observability](<https://devfeed.tech/topics/ai-observability.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-agent-data](<https://devfeed.tech/tags/ai-agent-data.md>), [ai-agent-observability](<https://devfeed.tech/tags/ai-agent-observability.md>), [ai-agent-trust](<https://devfeed.tech/tags/ai-agent-trust.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-reliability](<https://devfeed.tech/tags/ai-reliability.md>), [all](<https://devfeed.tech/tags/all.md>), [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [artificial-intelligence-agents](<https://devfeed.tech/tags/artificial-intelligence-agents.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-agents](<https://devfeed.tech/tags/autonomous-agents.md>), [away](<https://devfeed.tech/tags/away.md>), [barr-moses](<https://devfeed.tech/tags/barr-moses.md>), [bayes](<https://devfeed.tech/tags/bayes.md>), [big-data](<https://devfeed.tech/tags/big-data.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [data](<https://devfeed.tech/tags/data.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [enterprises](<https://devfeed.tech/tags/enterprises.md>), [faster](<https://devfeed.tech/tags/faster.md>), [gotten](<https://devfeed.tech/tags/gotten.md>), [launches](<https://devfeed.tech/tags/launches.md>), [monitoring](<https://devfeed.tech/tags/monitoring.md>), [monte-carlo](<https://devfeed.tech/tags/monte-carlo.md>), [news](<https://devfeed.tech/tags/news.md>), [reliable-ai-agents](<https://devfeed.tech/tags/reliable-ai-agents.md>), [self-improving-ai-agents](<https://devfeed.tech/tags/self-improving-ai-agents.md>), [the-latest](<https://devfeed.tech/tags/the-latest.md>), [top-story-5](<https://devfeed.tech/tags/top-story-5.md>)

### AI overview

Monte Carlo Data's Bayes is an AI agent designed to monitor other agents and the data environments they depend on. It discovers deployed agents, checks their outputs, traces problems through the model and data pipelines to pre-deployment decisions, and can triage or apply fixes autonomously before verifying operation. Customers can choose how much responsibility it takes; Bayes also integrates with third-party coding assistants and software deployment pipelines. Axios uses agents for newsroom tasks and says accuracy depends on correct input data and agent behavior. The article reports that one South American financial services institution saw Bayes detect 200% more incidents than its internal monitoring systems while reducing false positives by 80%; Snowflake uses it to help customers run Cortex-based agents. These are company-reported early-adopter results.

### Source excerpt

Enthusiasm for artificial intelligence agents is at an all-time high and many enterprises have gotten so carried away that they're deploying them faster than they keep up. Their human AI engineers only have the capacity to watch over a handful of AI agents at any one time, so they're in desperate need of help. What [...] The post Monte Carlo Data launches Bayes, an AI agent for autonomously monitoring and fixing autonomous agents appeared first on SiliconANGLE.

## How Sweetwater catches broken dbt pipelines before anyone notices, with Datadog

DevFeed: [How Sweetwater catches broken dbt pipelines before anyone notices, with Datadog](<https://devfeed.tech/articles/how-sweetwater-catches-broken-dbt-pipelines-before-anyone-notices-with-datadog-102654.md>)

Original publisher: [Read original article](<https://www.getdbt.com/blog/sweetwater-dbt-pipeline-monitoring-datadog>)

Author: Daniel Poppy

Published: 2026-10-27T18:08:00Z

Content type: news

Language: en

Sources: [dbt Labs](<https://devfeed.tech/sources/dbt-labs.md>)

Topics: [data observability](<https://devfeed.tech/topics/data-observability.md>), [datadog](<https://devfeed.tech/topics/datadog.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [Apache Kafka](<https://devfeed.tech/topics/apache-kafka.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [alert](<https://devfeed.tech/tags/alert.md>), [alerting](<https://devfeed.tech/tags/alerting.md>), [ci-cd](<https://devfeed.tech/tags/ci-cd.md>), [cli](<https://devfeed.tech/tags/cli.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [dag](<https://devfeed.tech/tags/dag.md>), [dashboard](<https://devfeed.tech/tags/dashboard.md>), [dashboards](<https://devfeed.tech/tags/dashboards.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [data-observability](<https://devfeed.tech/tags/data-observability.md>), [data-platform](<https://devfeed.tech/tags/data-platform.md>), [data-warehouse](<https://devfeed.tech/tags/data-warehouse.md>), [datadog](<https://devfeed.tech/tags/datadog.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [learn](<https://devfeed.tech/tags/learn.md>), [missing-data](<https://devfeed.tech/tags/missing-data.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>)

### AI overview

Sweetwater extended Datadog monitoring into its dbt pipelines as its data platform grew, building on the tool already used by IT to monitor source systems. In a recurring sales dashboard issue, data appeared fresh even though invoices lacked sales channels. Kafka insert dashboards helped the team trace a transaction spike to a websql source and an incorrectly scheduled legacy script; they rescheduled it and began considering a move to the cloud warehouse. Datadog's Google Cloud integrations and prebuilt metrics also helped the team manage timeout risk as DAGs and batch demands grew. Its setup includes schema- and database-wide monitors, table-adjusted row count checks, freshness checks that allow planned nightly downtime, team-specific alert and recovery routing, and machine-learning thresholds with manual annotations. The team is expanding into lineage and query monitoring, terminal and AI-assistant access through PupCLI and the Datadog MCP server, and CI/CD impact analysis and data profiling for changed models.

### Source excerpt

Sweetwater uses Datadog to catch broken dbt pipelines before dashboard users notice missing data.

## How Verily Health monitors wastewater for 30-plus pathogens with a 5-person team and 1 dbt package

DevFeed: [How Verily Health monitors wastewater for 30-plus pathogens with a 5-person team and 1 dbt package](<https://devfeed.tech/articles/how-verily-health-monitors-wastewater-for-30-plus-pathogens-with-a-5-person-team-and-1-dbt-package-101843.md>)

Original publisher: [Read original article](<https://www.getdbt.com/blog/verily-wastewater-pathogen-monitoring-dbt>)

Author: Daniel Poppy

Published: 2026-10-26T22:24:00Z

Content type: news

Language: en

Sources: [dbt Labs](<https://devfeed.tech/sources/dbt-labs.md>)

Topics: [Tool](<https://devfeed.tech/topics/tool.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [academic](<https://devfeed.tech/tags/academic.md>), [agencies](<https://devfeed.tech/tags/agencies.md>), [agent](<https://devfeed.tech/tags/agent.md>), [also](<https://devfeed.tech/tags/also.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [analytics-engineering](<https://devfeed.tech/tags/analytics-engineering.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [health](<https://devfeed.tech/tags/health.md>), [learn](<https://devfeed.tech/tags/learn.md>), [monitor](<https://devfeed.tech/tags/monitor.md>), [monitors](<https://devfeed.tech/tags/monitors.md>), [one](<https://devfeed.tech/tags/one.md>), [package](<https://devfeed.tech/tags/package.md>), [plus](<https://devfeed.tech/tags/plus.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [team](<https://devfeed.tech/tags/team.md>), [time](<https://devfeed.tech/tags/time.md>), [uses](<https://devfeed.tech/tags/uses.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

Verily Health's five-person data team uses a shared, parameterized dbt package to support near-real-time wastewater monitoring for more than 30 pathogens, while reusing the same FHIR resource macros across other health datasets. The package accommodates FHIR's nested data and lets the team adjust model sensitivity and grain through parameters instead of duplicating code. To keep models usable, the team limits composite keys to about three columns and most tables to roughly 30 columns, sometimes choosing a coarser grain. Its pipeline runs from a restricted bronze layer through FHIR-aligned staging and optional intermediate models to domain-focused marts. Freshness tests monitor both internal pipelines and partner data in varied formats; dbt's semantic layer and documentation help stakeholders analyze data without relying on the small team as a bottleneck. The excerpt reports a World Cup host-city monitoring partnership, with no major outbreaks reported during the tournament.

### Source excerpt

Verily Health's five-person team uses one dbt package to monitor wastewater for over 30 pathogens in near real time.

## Data Engineering Weekly #293

DevFeed: [Data Engineering Weekly #293](<https://devfeed.tech/articles/data-engineering-weekly-293-99730.md>)

Original publisher: [Read original article](<https://www.dataengineeringweekly.com/p/data-engineering-weekly-293>)

Author: Ananth Packkildurai

Published: 2026-10-26T02:12:10Z

Content type: news

Language: en

Sources: [Data Engineering Weekly](<https://devfeed.tech/sources/data-engineering-weekly.md>)

Topics: [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [semantic-layer](<https://devfeed.tech/topics/semantic-layer.md>), [metric-standardization](<https://devfeed.tech/topics/metric-standardization.md>), [Apache Kafka](<https://devfeed.tech/topics/apache-kafka.md>), [streaming data analytics](<https://devfeed.tech/topics/streaming-data-analytics.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Data science and engineering](<https://devfeed.tech/topics/data-science-and-engineering.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [about](<https://devfeed.tech/tags/about.md>), [adoption](<https://devfeed.tech/tags/adoption.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-native](<https://devfeed.tech/tags/ai-native.md>), [alert](<https://devfeed.tech/tags/alert.md>), [all](<https://devfeed.tech/tags/all.md>), [also](<https://devfeed.tech/tags/also.md>), [am](<https://devfeed.tech/tags/am.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [apache](<https://devfeed.tech/tags/apache.md>), [apple](<https://devfeed.tech/tags/apple.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [newsletter](<https://devfeed.tech/tags/newsletter.md>), [weekly](<https://devfeed.tech/tags/weekly.md>)

### AI overview

The newsletter highlights how shared definitions and abstractions can make data systems more reliable and adaptable. Pinterest's Metrics Board automates metric calculation, quality checks, and publishing from shared definitions; Meta's Analytics Libraries generate pipelines from reusable patterns, and an agent tool builds SQL from defined data models. LinkedIn separated competing event workloads, removed repeated request data, and shifted traffic gradually while comparing sampled outputs, reaching 90% without migration-related incidents. Snowflake describes batching Postgres changes through Iceberg and applying related updates together, with queries able to include pending changes. Meesho-Queue centralizes Kafka settings and assignments behind stable queue IDs, enabling migration of about 5,000 topics without application code changes. These are newsletter summaries of selected articles, not full coverage of them.

### Source excerpt

The Weekly Data Engineering Newsletter

## The Hidden Cost of Real-Time Data: Why Faster Pipelines Don't Always Deliver Better Decisions

DevFeed: [The Hidden Cost of Real-Time Data: Why Faster Pipelines Don't Always Deliver Better Decisions](<https://devfeed.tech/articles/the-hidden-cost-of-real-time-data-why-faster-pipelines-don-t-always-deliver-better-decisions-98772.md>)

Original publisher: [Read original article](<https://hackernoon.com/the-hidden-cost-of-real-time-data-why-faster-pipelines-dont-always-deliver-better-decisions?source=rss>)

Author: FNU Aditya Ranjan

Published: 2026-10-24T01:40:37Z

Content type: article

Language: en

Sources: [HackerNoon](<https://devfeed.tech/sources/hackernoon.md>)

Topics: [Event-Streaming](<https://devfeed.tech/topics/event-streaming.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [streaming data analytics](<https://devfeed.tech/topics/streaming-data-analytics.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>), [Apache Kafka](<https://devfeed.tech/topics/apache-kafka.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Cloud Dataflow](<https://devfeed.tech/topics/cloud-dataflow.md>)

Tags: [apache-flink](<https://devfeed.tech/tags/apache-flink.md>), [apache-kafka](<https://devfeed.tech/tags/apache-kafka.md>), [api](<https://devfeed.tech/tags/api.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [business](<https://devfeed.tech/tags/business.md>), [business-value](<https://devfeed.tech/tags/business-value.md>), [cdc](<https://devfeed.tech/tags/cdc.md>), [cost](<https://devfeed.tech/tags/cost.md>), [costs](<https://devfeed.tech/tags/costs.md>), [data](<https://devfeed.tech/tags/data.md>), [data-architecture](<https://devfeed.tech/tags/data-architecture.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-freshness](<https://devfeed.tech/tags/data-freshness.md>), [data-pipelines](<https://devfeed.tech/tags/data-pipelines.md>), [data-quality](<https://devfeed.tech/tags/data-quality.md>), [distributed-systems](<https://devfeed.tech/tags/distributed-systems.md>), [kafka](<https://devfeed.tech/tags/kafka.md>), [latency](<https://devfeed.tech/tags/latency.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [real-time](<https://devfeed.tech/tags/real-time.md>), [real-time-data](<https://devfeed.tech/tags/real-time-data.md>), [real-time-data-pipelines](<https://devfeed.tech/tags/real-time-data-pipelines.md>), [recovery](<https://devfeed.tech/tags/recovery.md>), [stream-processing](<https://devfeed.tech/tags/stream-processing.md>), [streaming](<https://devfeed.tech/tags/streaming.md>)

### AI overview

Fast event processing does not ensure timely, trustworthy information: end-to-end freshness, completeness, corrections, recovery, and operating cost all matter. Measure freshness from event time to consumer availability, since processing-time metrics can miss upstream and downstream delays. Exactly-once guarantees depend on framework, source, state, sink, and transaction boundaries; external actions may still need stable event IDs, idempotent writes, deduplication, or reconciliation. Late events trade immediate stability for completeness, so distinguish preliminary from finalized results and define correction windows. Stateful processing adds storage and recovery costs, requiring business-driven retention. The article proposes separate availability, completeness, correction, and recovery objectives, and evaluating designs under delayed, repeated, and failed-event scenarios using end-to-end latency, corrections, recovery, and cost.

### Source excerpt

Explore the hidden costs of real-time data pipelines and learn why latency, data freshness, correctness, and recovery matter in Kafka and streaming architecture

## Data Engineering Weekly #292

DevFeed: [Data Engineering Weekly #292](<https://devfeed.tech/articles/data-engineering-weekly-292-89969.md>)

Original publisher: [Read original article](<https://www.dataengineeringweekly.com/p/data-engineering-weekly-292>)

Author: Ananth Packkildurai

Published: 2026-10-19T03:58:54Z

Content type: article

Language: en

Sources: [Data Engineering Weekly](<https://devfeed.tech/sources/data-engineering-weekly.md>)

Topics: [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [RDF (Resource Description Framework)](<https://devfeed.tech/topics/rdf.md>), [Apache Flink](<https://devfeed.tech/topics/apache-flink.md>), [elasticsearch](<https://devfeed.tech/topics/elasticsearch.md>), [Amazon Bedrock Knowledge Bases](<https://devfeed.tech/topics/amazon-bedrock-knowledge-bases.md>), [deltalake](<https://devfeed.tech/topics/deltalake.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [airbnb](<https://devfeed.tech/tags/airbnb.md>), [apache](<https://devfeed.tech/tags/apache.md>), [apache-arrow](<https://devfeed.tech/tags/apache-arrow.md>), [aws](<https://devfeed.tech/tags/aws.md>), [aws-bedrock](<https://devfeed.tech/tags/aws-bedrock.md>), [clickhouse](<https://devfeed.tech/tags/clickhouse.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>)

### AI overview

This issue of Data Engineering Weekly collects reports and resources on data infrastructure. Topics include Zephon's deterministic online execution for AI training data, Apache Arrow's columnar computing ecosystem, observability ontology at Netflix, Spark and Flink shuffle scaling at LinkedIn, and data ingestion, search, and surveillance systems at other companies.

### Source excerpt

The Weekly Data Engineering Newsletter

## Data Engineering Weekly #291

DevFeed: [Data Engineering Weekly #291](<https://devfeed.tech/articles/data-engineering-weekly-291-85196.md>)

Original publisher: [Read original article](<https://www.dataengineeringweekly.com/p/data-engineering-weekly-291>)

Author: Ananth Packkildurai

Published: 2026-10-12T03:18:01Z

Content type: news

Language: en

Sources: [Data Engineering Weekly](<https://devfeed.tech/sources/data-engineering-weekly.md>)

Topics: [Platform Engineering](<https://devfeed.tech/topics/platform-engineering.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [Apache Kafka](<https://devfeed.tech/topics/apache-kafka.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Redpanda](<https://devfeed.tech/topics/redpanda-data-streaming.md>), [Amazon Bedrock Knowledge Bases](<https://devfeed.tech/topics/amazon-bedrock-knowledge-bases.md>), [engineering-culture](<https://devfeed.tech/topics/engineering-culture.md>)

Tags: [access-controls](<https://devfeed.tech/tags/access-controls.md>), [amazon-s3](<https://devfeed.tech/tags/amazon-s3.md>), [apache](<https://devfeed.tech/tags/apache.md>), [apache-spark](<https://devfeed.tech/tags/apache-spark.md>), [api](<https://devfeed.tech/tags/api.md>), [article](<https://devfeed.tech/tags/article.md>), [aws](<https://devfeed.tech/tags/aws.md>), [coinbase](<https://devfeed.tech/tags/coinbase.md>), [computing-capacity](<https://devfeed.tech/tags/computing-capacity.md>), [context-window](<https://devfeed.tech/tags/context-window.md>), [data](<https://devfeed.tech/tags/data.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [data-quality](<https://devfeed.tech/tags/data-quality.md>), [dynamodb](<https://devfeed.tech/tags/dynamodb.md>), [engineering](<https://devfeed.tech/tags/engineering.md>)

### AI overview

This issue of Data Engineering Weekly covers data platform design, a Booking.com data lake migration, Affirm's customer intelligence layer, blob-stream as an alternative to Kafka, LinkedIn's context layer for AI agents, Spark resource optimization, and LLM-generated user journeys. It also promotes data and AI events and sponsored guides.

### Source excerpt

The Weekly Data Engineering Newsletter

## dbt adds Redshift data sharing for cross-database writes and faster runs

DevFeed: [dbt adds Redshift data sharing for cross-database writes and faster runs](<https://devfeed.tech/articles/dbt-now-supports-amazon-redshift-data-sharing-and-your-pipelines-just-got-a-lot-faster-84273.md>)

Original publisher: [Read original article](<https://www.getdbt.com/blog/dbt-now-supports-amazon-redshift-data-sharing>)

Author: Stephen Robb

Published: 2026-10-09T22:02:00Z

Content type: release

Language: en

Sources: [dbt Labs](<https://devfeed.tech/sources/dbt-labs.md>)

Topics: [Amazon Redshift](<https://devfeed.tech/topics/amazon-redshift.md>), [Amazon S3](<https://devfeed.tech/topics/amazon-s3.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [AWS Organizations](<https://devfeed.tech/topics/aws-organizations.md>)

Tags: [amazon-redshift](<https://devfeed.tech/tags/amazon-redshift.md>), [aws](<https://devfeed.tech/tags/aws.md>), [build](<https://devfeed.tech/tags/build.md>), [ci](<https://devfeed.tech/tags/ci.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [database](<https://devfeed.tech/tags/database.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [partnerships](<https://devfeed.tech/tags/partnerships.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>)

### AI overview

dbt adds native Amazon Redshift data sharing support, allowing models to write across databases and teams to use shared data across clusters without copying it. An adapter metadata API upgrade also improves metadata resolution and run performance, including under high concurrency. The release supports isolated development and CI workflows, with limitations such as consumer-side views being unable to reference datashared tables directly.

### Source excerpt

dbt now supports Amazon Redshift data sharing: write across databases, isolate dev and CI, and run faster without copying data.

## How California built its Behavioral Health Public County Profile on Databricks

DevFeed: [How California built its Behavioral Health Public County Profile on Databricks](<https://devfeed.tech/articles/how-california-built-its-behavioral-health-public-county-profile-on-databricks-84226.md>)

Original publisher: [Read original article](<https://www.databricks.com/blog/how-california-built-its-behavioral-health-public-county-profile-databricks>)

Author: Michael Smith; Cameron Kahn; Stephen Hsu; Scott Black; Ben McKay; Varun Ganti

Published: 2026-10-09T21:30:00Z

Content type: article

Language: en

Sources: [Databricks](<https://devfeed.tech/sources/databricks.md>)

Topics: [databricks](<https://devfeed.tech/topics/databricks.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [product analytics](<https://devfeed.tech/topics/product-analytics.md>), [infrastructure monitoring](<https://devfeed.tech/topics/infrastructure-monitoring.md>)

Tags: [california](<https://devfeed.tech/tags/california.md>), [dashboard](<https://devfeed.tech/tags/dashboard.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-platform](<https://devfeed.tech/tags/data-platform.md>), [data-platforms](<https://devfeed.tech/tags/data-platforms.md>), [databricks](<https://devfeed.tech/tags/databricks.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [explore](<https://devfeed.tech/tags/explore.md>), [fedramp](<https://devfeed.tech/tags/fedramp.md>), [housing](<https://devfeed.tech/tags/housing.md>), [industries](<https://devfeed.tech/tags/industries.md>), [mental-health](<https://devfeed.tech/tags/mental-health.md>)

### AI overview

California's Department of Health Care Services built a public dashboard on Databricks that brings behavioral health data from all 58 counties together. The article describes how the platform supports data integration, governed analytics, and public access in a FedRAMP-authorized environment, helping residents and policymakers compare plans, spending, and outcomes.

### Source excerpt

The California Department of Health Care Services (DHCS) built the Behavioral Health...

## Data Engineering Is the Cache Invalidation Problem ft. Josh Wills

DevFeed: [Data Engineering Is the Cache Invalidation Problem ft. Josh Wills](<https://devfeed.tech/articles/data-engineering-is-the-cache-invalidation-problem-ft-josh-wills-84268.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/cache-invalidation-problem-data-engineering-josh-wills>)

Author: Simon Späti

Published: 2026-10-09T00:00:00Z

Content type: article

Language: en

Sources: [MotherDuck](<https://devfeed.tech/sources/motherduck.md>)

Topics: [data observability](<https://devfeed.tech/topics/data-observability.md>), [cache-invalidation](<https://devfeed.tech/topics/cache-invalidation.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [engineering-leadership](<https://devfeed.tech/topics/engineering-leadership.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>)

### AI overview

An interview with data engineer Josh Wills explores how AI agents affect large-scale data engineering. He says agents can handle routine tasks such as flaky tests and database migrations, but long-running Spark and Ray pipelines lack the fast feedback loops that help agents produce reliable, efficient work. Wills argues that architectural decisions, communicating trade-offs, and reviewing agent output remain central to engineering, and describes data engineering as the cache invalidation problem in a batch-processing context.

### Source excerpt

Josh Wills, creator of dbt-duckdb and former head of data engineering at Slack, on why the model barely matters, why agents still write ugly and expensive data pipelines, and why data engineering is still the cache invalidation problem.

## Introducing job-inspector: dltHub's first verified agent for diagnosing failed pipelines

DevFeed: [Introducing job-inspector: dltHub's first verified agent for diagnosing failed pipelines](<https://devfeed.tech/articles/introducing-job-inspector-dlthub-s-first-verified-agent-for-diagnosing-failed-pipelines-80337.md>)

Original publisher: [Read original article](<https://dlthub.com/blog/introducing-background-agents>)

Author: Elisabeth Reitmayr

Published: 2026-10-08T10:00:00Z

Content type: release

Language: en

Sources: [DLT Hub](<https://devfeed.tech/sources/dlt-hub.md>)

Topics: [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [all](<https://devfeed.tech/tags/all.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [data-engineering](<https://devfeed.tech/tags/data-engineering.md>), [data-quality](<https://devfeed.tech/tags/data-quality.md>), [debugging](<https://devfeed.tech/tags/debugging.md>), [diagnose](<https://devfeed.tech/tags/diagnose.md>), [dispatch](<https://devfeed.tech/tags/dispatch.md>), [docs](<https://devfeed.tech/tags/docs.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [experts](<https://devfeed.tech/tags/experts.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [pipelines](<https://devfeed.tech/tags/pipelines.md>), [product](<https://devfeed.tech/tags/product.md>), [slack](<https://devfeed.tech/tags/slack.md>), [updates](<https://devfeed.tech/tags/updates.md>)

### AI overview

dltHub has released job-inspector in public preview as its first verified background agent. It diagnoses failed data pipeline jobs using run data, logs and job definitions, then reports a likely cause, confidence and supporting evidence to Slack. It operates with read-only access, while engineers review the diagnosis and decide on production fixes.

### Source excerpt

dltHub background agents are in public preview, starting with job-inspector: a verified agent that diagnoses failed pipeline jobs with read-only access and posts the cause, its confidence and the log evidence to Slack.

[Next page](<https://devfeed.tech/topics/data-observability.md?cursor=WyIyMDI2LTEwLTA4VDEwOjAwOjAwKzAwOjAwIiwgImQ0Njc4ODJmLTFkZGUtNDliYS04Mzg4LTU2NWM4ZjljY2M0YSJd>)