# MotherDuck

Welcome to MotherDuck's blog posts!

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

## MotherDuck, Meet Excel

DevFeed: [MotherDuck, Meet Excel](<https://devfeed.tech/articles/motherduck-meet-excel-146538.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/motherduck-duckdb-meet-excel>)

Author: Mehdi Ouazza; Luka Lovosevic

Published: 2026-11-19T00:00:00Z

Content type: release

Language: en

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

Topics: [rdbms](<https://devfeed.tech/topics/rdbms.md>), [data transformation](<https://devfeed.tech/topics/data-transformation.md>)

Tags: [excel](<https://devfeed.tech/tags/excel.md>), [experimental](<https://devfeed.tech/tags/experimental.md>), [first](<https://devfeed.tech/tags/first.md>), [load](<https://devfeed.tech/tags/load.md>), [mac](<https://devfeed.tech/tags/mac.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [preview](<https://devfeed.tech/tags/preview.md>), [refresh](<https://devfeed.tech/tags/refresh.md>), [web](<https://devfeed.tech/tags/web.md>), [windows](<https://devfeed.tech/tags/windows.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

MotherDuck for Excel is an experimental add-in for Mac, Windows, and the web that connects Excel workbooks to MotherDuck. It loads tables or server-aggregated PivotTables, supports one-click refresh, and lets users filter, sort, and limit rows in MotherDuck before they reach Excel. Filters can follow worksheet cells; users can also run parameterized SQL, request SQL from MotherDuck's prompt_sql and review it before loading, or use MOTHERDUCK.TABLE and MOTHERDUCK.SQL formulas. Saved imports retain queries and filters but not credentials, so workbook recipients refresh with their own access. The Write back tab can create, replace, append to, or update/insert into a MotherDuck table from a range or Excel Table; a preview shows proposed changes before anything is applied. This is a Labs preview that may change or break, and write-back modifies real tables, so the source advises trying it on reproducible data.

### Source excerpt

MotherDuck for Excel is an experimental add-in for Excel on Mac, Windows and the web. Load tables and PivotTables, refresh them with one click, and write ranges back to MotherDuck with a preview first.

## Business Knowledge Is Still King: Agentic Data Analysis Without a Data Team

DevFeed: [Business Knowledge Is Still King: Agentic Data Analysis Without a Data Team](<https://devfeed.tech/articles/business-knowledge-is-still-king-agentic-data-analysis-without-a-data-team-138190.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/business-knowledge-agentic-data-analysis-without-data-team>)

Author: Simon Späti

Published: 2026-11-13T00:00:00Z

Content type: article

Language: en

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

Topics: [business-intelligence](<https://devfeed.tech/topics/business-intelligence.md>), [Data analysis](<https://devfeed.tech/topics/data-analysis.md>), [bi-tools](<https://devfeed.tech/topics/bi-tools.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [etl](<https://devfeed.tech/topics/etl.md>), [Data visualization](<https://devfeed.tech/topics/data-visualization.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [builds](<https://devfeed.tech/tags/builds.md>), [business](<https://devfeed.tech/tags/business.md>), [claude](<https://devfeed.tech/tags/claude.md>), [dashboard](<https://devfeed.tech/tags/dashboard.md>), [data](<https://devfeed.tech/tags/data.md>), [data-analysis](<https://devfeed.tech/tags/data-analysis.md>), [data-team](<https://devfeed.tech/tags/data-team.md>), [domain](<https://devfeed.tech/tags/domain.md>), [knowledge](<https://devfeed.tech/tags/knowledge.md>), [live](<https://devfeed.tech/tags/live.md>), [map](<https://devfeed.tech/tags/map.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [numbers](<https://devfeed.tech/tags/numbers.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [public](<https://devfeed.tech/tags/public.md>), [right](<https://devfeed.tech/tags/right.md>), [sources](<https://devfeed.tech/tags/sources.md>), [team](<https://devfeed.tech/tags/team.md>)

### AI overview

The article's central finding is that AI agents can help a domain expert without SQL skills build and explore a data pipeline and dashboard, but business knowledge and human verification remain essential. In its MotherDuck example, Claude combines MeteoSwiss weather data, Swissgrid electricity data, and canton boundaries to estimate a Swiss electricity heating threshold, create a map, and assemble a shareable dashboard. Follow-up checks exposed a 2025 timestamp parsing bug that had dropped 61% of rows, illustrating why results need scrutiny; the article also says 2025 and 2026 estimates differed by temperature reference and warns that agents can use excessive permissions, run costly queries, mishandle sensitive data, or create hard-to-maintain stacks. The author presents MotherDuck's Flights, Dives, and Guides as an integrated route to scheduled, shareable analysis, while noting that mature automation, governance, permissions, and organizational integration still need data-team ownership. Alternatives mentioned include local DuckDB with Claude, Hex, and agents in existing warehouse platforms, each with different sharing, scheduling, or governance tradeoffs.

### Source excerpt

A domain expert with no SQL skills builds a working pipeline, map and live dashboard from three public sources using Claude and MotherDuck, and why business knowledge still decides if the numbers are right.

## The dawn of the Data Intelligence Engineer

DevFeed: [The dawn of the Data Intelligence Engineer](<https://devfeed.tech/articles/the-dawn-of-the-data-intelligence-engineer-130712.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/the-data-intelligence-engineer>)

Author: Jordan Tigani

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

Content type: opinion

Language: en

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

Topics: [AI for data automation](<https://devfeed.tech/topics/ai-for-data-automation.md>), [business-intelligence](<https://devfeed.tech/topics/business-intelligence.md>), [Data Management](<https://devfeed.tech/topics/data-management.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [bi-tools](<https://devfeed.tech/topics/bi-tools.md>), [DataOps](<https://devfeed.tech/topics/dataops.md>)

Tags: [2025](<https://devfeed.tech/tags/2025.md>), [about](<https://devfeed.tech/tags/about.md>), [accountability](<https://devfeed.tech/tags/accountability.md>), [act](<https://devfeed.tech/tags/act.md>), [admin](<https://devfeed.tech/tags/admin.md>), [age-of-ai](<https://devfeed.tech/tags/age-of-ai.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-evals](<https://devfeed.tech/tags/agent-evals.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [can](<https://devfeed.tech/tags/can.md>), [data](<https://devfeed.tech/tags/data.md>), [data-teams](<https://devfeed.tech/tags/data-teams.md>), [engineer](<https://devfeed.tech/tags/engineer.md>), [intelligence](<https://devfeed.tech/tags/intelligence.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [next](<https://devfeed.tech/tags/next.md>), [teams](<https://devfeed.tech/tags/teams.md>), [this](<https://devfeed.tech/tags/this.md>), [work](<https://devfeed.tech/tags/work.md>)

### AI overview

The proposed Data Intelligence Engineer shifts data work from producing every query and dashboard to making agent-driven self-service reliable. The role maintains clear data models and shared business context, builds frameworks, monitors systems, and uses known-answer evaluations to check agent results. Human accountability remains essential because agents may misinterpret company-specific definitions or select the wrong data. The author recommends supervision proportional to an output's impact: the data team can leave low-stakes experiments largely open, make a best effort to fix important curated dashboards and apps, and keep core metrics highly reliable. In the author's MotherDuck example, Flights handle ingestion pipelines, Guides provide shared model and business context, Dives are agent-created visualizations, and each user or agent works in a dedicated DuckDB instance called a Duckling; these are reachable through SQL to reduce tool handoffs. The excerpt presents self-service as a way for stakeholders to ask follow-up questions and make timely decisions, while acknowledging the risk of more people producing errors.

### Source excerpt

For Data Teams, AI has changed everything. The traditional work can be run by agents, leaving folks wondering "what's next"? In this blog, MotherDuck CEO Jordan Tigani unpacks the possibilities.

## Introducing Agent Chat: The Shortest Path from Data to Action

DevFeed: [Introducing Agent Chat: The Shortest Path from Data to Action](<https://devfeed.tech/articles/introducing-agent-chat-the-shortest-path-from-data-to-action-116322.md>)

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

Author: Adithya Krishnan; Sheila Sitaram

Published: 2026-11-03T00:00:00Z

Content type: release

Language: en

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

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [AI for data automation](<https://devfeed.tech/topics/ai-for-data-automation.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [analytics stack](<https://devfeed.tech/topics/analytics-stack.md>), [Data Science](<https://devfeed.tech/topics/data-science.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-chat](<https://devfeed.tech/tags/ai-chat.md>), [ai-native](<https://devfeed.tech/tags/ai-native.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [analytics-infrastructure](<https://devfeed.tech/tags/analytics-infrastructure.md>), [analyze](<https://devfeed.tech/tags/analyze.md>), [app](<https://devfeed.tech/tags/app.md>), [applications](<https://devfeed.tech/tags/applications.md>), [chat](<https://devfeed.tech/tags/chat.md>), [context](<https://devfeed.tech/tags/context.md>), [data](<https://devfeed.tech/tags/data.md>), [insights](<https://devfeed.tech/tags/insights.md>), [time](<https://devfeed.tech/tags/time.md>), [tools](<https://devfeed.tech/tags/tools.md>), [users](<https://devfeed.tech/tags/users.md>), [workflows](<https://devfeed.tech/tags/workflows.md>)

### AI overview

MotherDuck Agent Chat brings data exploration and workflow building into one conversational interface: its native agent can inspect datasets, write and run SQL or Python, create and schedule Flights for ingestion and transformation, and build shareable interactive Dive apps. Guides provide business context such as metric definitions and join relationships. The excerpt's example uses NOAA GHCN-Daily data on S3 to build a temperature analysis pipeline and an app for exploring trends by country and year. Users can continue work in the same conversation and revisit or share outputs; the article says data, workflows and apps remain in MotherDuck. It also cautions that agents can make incorrect assumptions and that generated outputs are not automatically trustworthy, describing centralized context and inspectable workflows as ways to help teams verify and reuse results. The article says users need not configure a separate AI client or choose a model; model optimization is handled by MotherDuck.

### Source excerpt

Analyze data, automate workflows, and turn insights into applications so users can act on insights without losing time and context switching between tools

## 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.

## Introducing Dive Apps: Superpowers for AI Builders

DevFeed: [Introducing Dive Apps: Superpowers for AI Builders](<https://devfeed.tech/articles/introducing-dive-apps-superpowers-for-ai-builders-107248.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/dive-apps-superpowers-for-ai-builders>)

Author: Alex Monahan

Published: 2026-10-29T00:00:00Z

Content type: release

Language: en

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

Topics: [CASE](<https://devfeed.tech/topics/casejs.md>), [human in the loop agent](<https://devfeed.tech/topics/human-in-the-loop-agent.md>), [etl](<https://devfeed.tech/topics/etl.md>), [AI for customer service automation](<https://devfeed.tech/topics/ai-for-customer-service-automation.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [Tool](<https://devfeed.tech/topics/tool.md>)

Tags: [000](<https://devfeed.tech/tags/000.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [apps](<https://devfeed.tech/tags/apps.md>), [automate](<https://devfeed.tech/tags/automate.md>), [build](<https://devfeed.tech/tags/build.md>), [can](<https://devfeed.tech/tags/can.md>), [custom](<https://devfeed.tech/tags/custom.md>), [data](<https://devfeed.tech/tags/data.md>), [data-warehouse](<https://devfeed.tech/tags/data-warehouse.md>), [dive](<https://devfeed.tech/tags/dive.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [operations](<https://devfeed.tech/tags/operations.md>), [run](<https://devfeed.tech/tags/run.md>), [saas](<https://devfeed.tech/tags/saas.md>), [superpowers](<https://devfeed.tech/tags/superpowers.md>), [warehouse](<https://devfeed.tech/tags/warehouse.md>), [with](<https://devfeed.tech/tags/with.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

Dive Apps turn MotherDuck visualizations into interactive tools that can trigger actions. A visualization can launch a serverless Python Flight to update MotherDuck or write to external systems, send messages, rerun models, or start an agent workflow. The article proposes custom apps for replacing selected SaaS features, supporting decisions, and automating operations. For complex processes, it recommends moving incrementally from advice to approval, exception handling, and then autonomy, with human review and write-back available during the transition. MotherDuck's stated advantages include centralized data access, on-demand analytical compute, per-user resources, SSO and role- or data-level access controls, and Guides for sharing data context across AI models. The examples and benefits are presented by the product's maker; the text does not provide independent performance evidence.

### Source excerpt

MotherDuck is launching Dive Apps: fully custom data apps that can write-back anywhere. Build them with your agent and they run on the MotherDuck data warehouse. Use Dive Apps to replace SaaS apps, automate your operations, or bring humans into the loop to steer your agents.

## Bringing the Big Ducks

DevFeed: [Bringing the Big Ducks](<https://devfeed.tech/articles/bringing-the-big-ducks-105858.md>)

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

Author: Jordan Tigani

Published: 2026-10-28T00:00:00Z

Content type: article

Language: en

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

Topics: [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [big-data](<https://devfeed.tech/topics/big-data.md>), [Query (disambiguation)](<https://devfeed.tech/topics/query.md>), [Ground truth / benchmark quality](<https://devfeed.tech/topics/ground-truth-benchmark-quality.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [amazon](<https://devfeed.tech/tags/amazon.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [architectures](<https://devfeed.tech/tags/architectures.md>), [arithmetic](<https://devfeed.tech/tags/arithmetic.md>), [azure](<https://devfeed.tech/tags/azure.md>), [azure-synapse](<https://devfeed.tech/tags/azure-synapse.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [bigquery](<https://devfeed.tech/tags/bigquery.md>), [claude](<https://devfeed.tech/tags/claude.md>), [client](<https://devfeed.tech/tags/client.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [cpus](<https://devfeed.tech/tags/cpus.md>), [data-warehouse](<https://devfeed.tech/tags/data-warehouse.md>), [database](<https://devfeed.tech/tags/database.md>), [databases](<https://devfeed.tech/tags/databases.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>)

### AI overview

MotherDuck's results suggest its single-node Giga instance can handle TPC-DS workloads at 1 TB and 10 TB scale while remaining competitive with distributed warehouses. On DuckDB 1.5.6, Giga's geometric mean query time was 690 ms at 1 TB and 5.6 seconds at 10 TB; a smaller Mega instance took 2 seconds at 1 TB. The 10 TB benchmark scans 168 TB across 99 queries. Comparisons use vendor or third-party results published in 2022, so they do not establish current speed rankings. MotherDuck ran unmodified DuckDB TPC-DS queries, reporting the best of three iterations per query and the geometric mean of hot query times. The author says some workloads still struggle to scale up and may need distributed support; DuckDB 2.0 improvements were not included.

### Source excerpt

The pursuit of all things duck-related has led us to some interesting places. In this blog, MotherDuck CEO Jordan Tigani explores the bigger, darker corners of benchmarks on top DuckDB 1.5 (and teases what is possible in DuckDB 2.0).

## DuckDB Ecosystem Newsletter : October 2026

DevFeed: [DuckDB Ecosystem Newsletter : October 2026](<https://devfeed.tech/articles/duckdb-ecosystem-newsletter-october-2026-90280.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/duckdb-ecosystem-newsletter-october-2026>)

Author: Simon Späti

Published: 2026-10-18T00:00:00Z

Content type: article

Language: en

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

Topics: [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [Parser](<https://devfeed.tech/topics/parser.md>), [async](<https://devfeed.tech/topics/async.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [app](<https://devfeed.tech/tags/app.md>), [async](<https://devfeed.tech/tags/async.md>), [browser](<https://devfeed.tech/tags/browser.md>), [community](<https://devfeed.tech/tags/community.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [ducklake](<https://devfeed.tech/tags/ducklake.md>), [ecosystem](<https://devfeed.tech/tags/ecosystem.md>), [jev](<https://devfeed.tech/tags/jev.md>), [mac](<https://devfeed.tech/tags/mac.md>), [maintenance](<https://devfeed.tech/tags/maintenance.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [native](<https://devfeed.tech/tags/native.md>), [newsletter](<https://devfeed.tech/tags/newsletter.md>), [october](<https://devfeed.tech/tags/october.md>), [october-2026](<https://devfeed.tech/tags/october-2026.md>), [openstreetmap](<https://devfeed.tech/tags/openstreetmap.md>), [peg](<https://devfeed.tech/tags/peg.md>)

### AI overview

The excerpt highlights DuckDB's expanding use in SQL, storage, browser and desktop workflows. Jev brings typed AI answers into DuckDB and MotherDuck SQL; cited examples report 89% accuracy on 100,000 AG News rows in 40 seconds for $0.50, while a regex remained faster for one classification task. DuckDB 2.0's PEG parser supports runtime grammar extensions and sharply reduces parsing time for malformed queries; its asynchronous S3 reads cut one 10,000-file JSON workload from about 297 to 99 seconds. DuckLake maintenance guidance pairs snapshot expiration with cleanup of old and orphaned S3 files. Other examples include spatially sorted OpenStreetMap Parquet data that enabled a 657-million-row count query with 2.2 MiB of bandwidth, and a browser-based OneLake SQL app using delegated Azure access, range requests and caching, with a 200,000-row result cap. The excerpt also describes local DuckDB-WASM file viewing and a native Mac client designed to render large Parquet files with low memory overhead.

### Source excerpt

The October 2026 DuckDB Ecosystem Newsletter: Jev typed AI answers in DuckDB and MotherDuck SQL, DuckDB 2.0's PEG parser and async S3 reads, DuckLake maintenance, analysis-ready OpenStreetMap, OneLake in the browser, a native Mac app, plus a community spotlight on Mark Litwintschik.

## prompt\_jev() now works on images

DevFeed: [prompt\_jev() now works on images](<https://devfeed.tech/articles/prompt-jev-now-works-on-images-88769.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/motherduck-announce-jev-image>)

Author: MotherDuck team

Published: 2026-10-16T00:00:00Z

Content type: release

Language: en

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

Topics: [multimodal-ai](<https://devfeed.tech/topics/multimodal-ai.md>), [Image](<https://devfeed.tech/topics/image.md>), [Prompt Engineering](<https://devfeed.tech/topics/prompt-engineering.md>), [Image processing](<https://devfeed.tech/topics/image-processing.md>)

Tags: [benchmark](<https://devfeed.tech/tags/benchmark.md>), [blob](<https://devfeed.tech/tags/blob.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [confidence-score](<https://devfeed.tech/tags/confidence-score.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [docs](<https://devfeed.tech/tags/docs.md>), [features](<https://devfeed.tech/tags/features.md>), [filter](<https://devfeed.tech/tags/filter.md>), [flight](<https://devfeed.tech/tags/flight.md>), [food](<https://devfeed.tech/tags/food.md>), [hugging-face](<https://devfeed.tech/tags/hugging-face.md>), [image](<https://devfeed.tech/tags/image.md>), [images](<https://devfeed.tech/tags/images.md>), [jev](<https://devfeed.tech/tags/jev.md>), [join](<https://devfeed.tech/tags/join.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [sql](<https://devfeed.tech/tags/sql.md>)

### AI overview

TypeSafe Jev 1.14 extends MotherDuck's `prompt_jev` to images stored as BLOBs: SQL questions return typed answers that can be filtered, grouped, and joined with other data. The excerpt shows reading image bytes from Hugging Face Parquet or loading S3 files into a table; Jev requires each image dimension to be 280-1,568 pixels, with resizing needed for larger images and smaller ones skipped in the example workflow. On 1,010 Food-101 images, keeping Jev answers with confidence at least 0.8 yielded 97.7% accuracy while covering 80% of photos. The excerpt also describes applying 14 image questions to about 77,000 Met works and grouping results with catalog dates to estimate when depicted smiles became more common. It notes Jev was not the top-performing model in a hot-dog comparison, while presenting confidence scores as a way to select answers.

### Source excerpt

prompt_jev() now takes images: store them as BLOBs in MotherDuck, ask closed questions in SQL, and get typed answers with a confidence score you can join with the rest of your data.

## MotherDuck is coming to Azure

DevFeed: [MotherDuck is coming to Azure](<https://devfeed.tech/articles/motherduck-is-coming-to-azure-88767.md>)

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

Author: Jacob Matson

Published: 2026-10-15T00:00:00Z

Content type: release

Language: en

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

Topics: [Azure](<https://devfeed.tech/topics/azure.md>), [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [Calculate Pi](<https://devfeed.tech/topics/calculate-pi.md>), [Visual Studio Code](<https://devfeed.tech/topics/visual-studio-code.md>), [Azure OpenAI](<https://devfeed.tech/topics/azure-openai.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [azure](<https://devfeed.tech/tags/azure.md>), [chatgpt](<https://devfeed.tech/tags/chatgpt.md>), [claude-code](<https://devfeed.tech/tags/claude-code.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [code](<https://devfeed.tech/tags/code.md>), [copilot](<https://devfeed.tech/tags/copilot.md>), [csv](<https://devfeed.tech/tags/csv.md>), [cursor](<https://devfeed.tech/tags/cursor.md>), [databases](<https://devfeed.tech/tags/databases.md>), [entra-id](<https://devfeed.tech/tags/entra-id.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>)

### AI overview

MotherDuck announces that it is building an Azure version of its serverless data warehouse, with Azure region availability to be guided by waitlist demand. The article describes its per-user Duckling compute model, integrations with Microsoft identity, Copilot and Power BI, and agent-oriented tools for querying and building data apps. Azure Blob Storage access is already available from MotherDuck on AWS.

### Source excerpt

MotherDuck is coming to Azure. Pick your region on the waitlist; regions open in order of demand. Your Copilot agents can connect to MotherDuck today.

## A Chief of Staff's Guide to Context Layers

DevFeed: [A Chief of Staff's Guide to Context Layers](<https://devfeed.tech/articles/a-chief-of-staff-s-guide-to-context-layers-88768.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/chief-of-staff-guide-ai-agents-metrics>)

Author: Kaitlyn Henry

Published: 2026-10-15T00:00:00Z

Content type: article

Language: en

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

Topics: [semantic-layer](<https://devfeed.tech/topics/semantic-layer.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>), [knowledge-engineering](<https://devfeed.tech/topics/knowledge-engineering.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [context](<https://devfeed.tech/topics/context.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [bi](<https://devfeed.tech/tags/bi.md>), [bootstrap](<https://devfeed.tech/tags/bootstrap.md>), [build](<https://devfeed.tech/tags/build.md>), [canonical](<https://devfeed.tech/tags/canonical.md>), [churn](<https://devfeed.tech/tags/churn.md>), [consensus](<https://devfeed.tech/tags/consensus.md>), [context](<https://devfeed.tech/tags/context.md>), [context-layer](<https://devfeed.tech/tags/context-layer.md>), [fragments](<https://devfeed.tech/tags/fragments.md>), [guide](<https://devfeed.tech/tags/guide.md>), [layers](<https://devfeed.tech/tags/layers.md>), [metrics](<https://devfeed.tech/tags/metrics.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [non-technical](<https://devfeed.tech/tags/non-technical.md>), [self-serve-analytics](<https://devfeed.tech/tags/self-serve-analytics.md>)

### AI overview

AI-enabled self-serve analytics makes warehouse access easier, but can leave teams with conflicting metric definitions. The author recommends having a small cross-functional group agree on key metrics, using an LLM to draft a data dictionary from the schema, then having domain experts document the missing business rules and pressure-test sample results with people close to the data. After a few review rounds, humans own the definition while the LLM generates SQL. The article argues that definitions should be stored near raw data because agents may query warehouses directly instead of routing through BI or modeling layers. It presents MotherDuck Guides as markdown context loaded through its MCP: shared admin Guides can take precedence over personal ones, and saves are versioned with change comments. The author's proposed balance is to make metric definitions, filters, and grain consistent while allowing flexibility in how agents write queries.

### Source excerpt

Kaitlyn, MotherDuck's Chief of Staff, on why self-serve analytics with AI agents fragments your metrics, and how non-technical leaders can get everyone on one definition of metrics.

## 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.

## 5 Jev use cases for analytics, with real queries and datasets

DevFeed: [5 Jev use cases for analytics, with real queries and datasets](<https://devfeed.tech/articles/5-jev-use-cases-for-analytics-with-real-queries-and-datasets-78939.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/jev-analytics-use-cases>)

Author: Mehdi Ouazza

Published: 2026-10-06T00:00:00Z

Content type: tutorial

Language: en

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

Topics: [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [managing large-scale text data in AI](<https://devfeed.tech/topics/managing-large-scale-text-data-in-ai.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-labs](<https://devfeed.tech/tags/ai-labs.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [alibaba-cloud](<https://devfeed.tech/tags/alibaba-cloud.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [data](<https://devfeed.tech/tags/data.md>), [datasets](<https://devfeed.tech/tags/datasets.md>), [hacker-news](<https://devfeed.tech/tags/hacker-news.md>), [jev](<https://devfeed.tech/tags/jev.md>), [job](<https://devfeed.tech/tags/job.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [queries](<https://devfeed.tech/tags/queries.md>), [regex](<https://devfeed.tech/tags/regex.md>), [sales](<https://devfeed.tech/tags/sales.md>), [sql](<https://devfeed.tech/tags/sql.md>), [traces](<https://devfeed.tech/tags/traces.md>), [use-cases](<https://devfeed.tech/tags/use-cases.md>)

### AI overview

The excerpt presents `prompt_jev()` as a MotherDuck SQL function for classifying text into predefined labels, yes/no answers, or scores with confidence. Examples cover judging agent traces, filtering Hacker News mentions and sentiment, screening sales calls, identifying primary cloud platforms in job postings, and classifying work modes. The article describes using explicit instructions, confidence thresholds, sampling, and human review; for cloud analysis it also uses escape-hatch labels and filters out vendor postings and ambiguous results. On 2,000 postings, `prompt_jev()` is reported about 20 times faster than a standard LLM call, while regex was fastest but required patches to handle false positives. The recommended approach is to use regex for structured patterns and Jev for meaning, validate thresholds on labeled examples, and materialize results to avoid repeat classification. Jev returns bounded choices rather than summaries or new categories; the excerpt also cautions that Hacker News is not representative market-share data and that customer-data handling rules should be checked.

### Source excerpt

Five prompt_jev() use cases on real data, with SQL you can rerun: AI agent traces, Hacker News sentiment, sales calls, cloud job postings and a regex replacement.

## Build It Yourself: What You'll Make at the Data Outpost Workshops

DevFeed: [Build It Yourself: What You'll Make at the Data Outpost Workshops](<https://devfeed.tech/articles/build-it-yourself-what-you-ll-make-at-the-data-outpost-workshops-78934.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/data-outpost-2026-workshops>)

Author: Alex Monahan

Published: 2026-10-05T00:00:00Z

Content type: article

Language: en

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

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [Learning](<https://devfeed.tech/topics/learning.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [data observability](<https://devfeed.tech/topics/data-observability.md>), [Developer Events](<https://devfeed.tech/topics/developer-events.md>), [streaming data analytics](<https://devfeed.tech/topics/streaming-data-analytics.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [airbyte](<https://devfeed.tech/tags/airbyte.md>), [braintrust](<https://devfeed.tech/tags/braintrust.md>), [build](<https://devfeed.tech/tags/build.md>), [data](<https://devfeed.tech/tags/data.md>), [estuary](<https://devfeed.tech/tags/estuary.md>), [here](<https://devfeed.tech/tags/here.md>), [here-s](<https://devfeed.tech/tags/here-s.md>), [make](<https://devfeed.tech/tags/make.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [need](<https://devfeed.tech/tags/need.md>), [one](<https://devfeed.tech/tags/one.md>), [production](<https://devfeed.tech/tags/production.md>), [run](<https://devfeed.tech/tags/run.md>), [san-francisco](<https://devfeed.tech/tags/san-francisco.md>), [sigma](<https://devfeed.tech/tags/sigma.md>), [with](<https://devfeed.tech/tags/with.md>), [workshops](<https://devfeed.tech/tags/workshops.md>), [you-ll](<https://devfeed.tech/tags/you-ll.md>)

### AI overview

Data Outpost's November 4, 2026 workshops are hands-on sessions in San Francisco, scheduled in three parallel time slots, with participants choosing one per slot and typically building a working project. The described projects range from enterprise-connected analytics agents, MotherDuck data stacks and Sigma workbooks to hosted MCP servers, warehouse migration runbooks, Iceberg and DuckLake examples, live database replication, Airbyte ingestion, and document pipelines. Follow-on sessions address evaluating production traces and understanding the model, harness, context, tools, memory, and inference components of long-running coding agents. The article emphasizes that agents depend on well-prepared data and that production use raises concerns such as correctness, reproducibility, permissions, observability, and evaluation.

### Source excerpt

Data Outpost kicks off on November 4, 2026 in San Francisco with 11 hands-on workshops from MotherDuck, Vercel, Sigma, Prefect, Estuary, Airbyte, Reducto, Braintrust, and Together AI. Here's what you'll build in each one: your own agents, the data pipelines they need, and the skills to run them in production.

## AI Needs Data, Data Needs AI: What You'll Learn at Data Outpost

DevFeed: [AI Needs Data, Data Needs AI: What You'll Learn at Data Outpost](<https://devfeed.tech/articles/ai-needs-data-data-needs-ai-what-you-ll-learn-at-data-outpost-78933.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/data-outpost-2026-agenda>)

Author: Alex Monahan

Published: 2026-10-01T00:00:00Z

Content type: article

Language: en

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

Topics: [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [semantic-layer](<https://devfeed.tech/topics/semantic-layer.md>), [AI-Ready Data](<https://devfeed.tech/topics/ai-ready-data.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Self Hosted AI Tools](<https://devfeed.tech/topics/self-hosted-ai-tools.md>), [data](<https://devfeed.tech/topics/data.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agent-oversight](<https://devfeed.tech/tags/agent-oversight.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-answers](<https://devfeed.tech/tags/ai-answers.md>), [ai-assistant](<https://devfeed.tech/tags/ai-assistant.md>), [ai-mode](<https://devfeed.tech/tags/ai-mode.md>), [ai-systems](<https://devfeed.tech/tags/ai-systems.md>), [ai-tools](<https://devfeed.tech/tags/ai-tools.md>), [canva](<https://devfeed.tech/tags/canva.md>), [conference](<https://devfeed.tech/tags/conference.md>), [context](<https://devfeed.tech/tags/context.md>), [data](<https://devfeed.tech/tags/data.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [doordash](<https://devfeed.tech/tags/doordash.md>), [san-francisco](<https://devfeed.tech/tags/san-francisco.md>), [sql](<https://devfeed.tech/tags/sql.md>), [stripe](<https://devfeed.tech/tags/stripe.md>), [talks](<https://devfeed.tech/tags/talks.md>), [teams](<https://devfeed.tech/tags/teams.md>)

### AI overview

Data Outpost's November 5, 2026 conference focuses on the data foundations needed to make AI useful and dependable in business. The listed sessions cover agents that write to data and how teams oversee them; scaling AI under latency and cost constraints; exposing business meaning and testing when AI should answer or refuse; and combining operational data, context, permissions, and agent exploration. A Positron talk argues that reliable AI-written SQL depends on the authoring harness--schema browsing, completions, an interactive console, and a data explorer--as much as on the model. Other sessions address modeling document data before scaling AI extraction and deploying document intelligence within regulated environments, including privacy rules, data residency, and build-versus-buy choices. The November 4 workshops include building a data agent, scaling DuckLake and Iceberg for agent workloads, streaming data with change data capture, and building an MCP server. Some agenda topics are still forthcoming.

### Source excerpt

Data Outpost is an AI and data conference in San Francisco. Its main day, November 5, 2026, has 13 talks and panels from teams at Canva, Stripe, DoorDash, Posit, dbt and more. Here's what each session covers, from agents that write back to context layers, SQL harnesses, and AI for regulated data, and what you'll walk away with.

## Jev for analytics: turn text columns into gold, in plain SQL

DevFeed: [Jev for analytics: turn text columns into gold, in plain SQL](<https://devfeed.tech/articles/jev-for-analytics-turn-text-columns-into-gold-in-plain-sql-78940.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/jev-for-analytics>)

Author: Mehdi Ouazza

Published: 2026-09-29T00:00:00Z

Content type: tutorial

Language: en

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

Topics: [SQL](<https://devfeed.tech/topics/sql.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [Rockset](<https://devfeed.tech/topics/rockset.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [alert](<https://devfeed.tech/tags/alert.md>), [all](<https://devfeed.tech/tags/all.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [column](<https://devfeed.tech/tags/column.md>), [consumer](<https://devfeed.tech/tags/consumer.md>), [cost](<https://devfeed.tech/tags/cost.md>), [jev](<https://devfeed.tech/tags/jev.md>), [llm](<https://devfeed.tech/tags/llm.md>), [plus](<https://devfeed.tech/tags/plus.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [sql](<https://devfeed.tech/tags/sql.md>), [text](<https://devfeed.tech/tags/text.md>), [trust](<https://devfeed.tech/tags/trust.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

`prompt_jev()` turns free text into a typed choice from labels you supply, letting repeated classification run once and leaving later analysis to ordinary SQL. In the excerpt's 100,000 CFPB complaints, it classified seven requested-outcome categories in 82.2 seconds; the launch benchmark reports $0.50 per 100,000 rows versus $1.58 for gpt-5-nano and $37.58 for GPT-5.6 Terra. The labels must be constant choices, though a DuckDB variable can hold a taxonomy stored in a table. Its confidence scores can flag uncertain rows for review, but high confidence does not guarantee correctness; a sample comparison also showed disagreement between models around fuzzy category boundaries. Use an LLM to develop categories or generate explanations, and test Jev on a sample before scaling. The excerpt says text is sent to TypeSafe for inference and advises checking data-handling requirements for sensitive data.

### Source excerpt

Turn a free-text column into typed SQL columns with prompt_jev(): 100,000 consumer complaints classified in 82 seconds, at about 1% of a frontier LLM's cost, plus how to check the labels before you trust them.

## Querying the entire internet (100 billion rows!) with MotherDuck

DevFeed: [Querying the entire internet (100 billion rows!) with MotherDuck](<https://devfeed.tech/articles/querying-the-entire-internet-100-billion-rows-with-motherduck-78943.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/querying-the-entire-internet-100-billion-rows-with-motherduck>)

Author: Dumky de Wilde

Published: 2026-09-22T00:00:00Z

Content type: article

Language: en

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

Topics: [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [hosting](<https://devfeed.tech/topics/hosting.md>), [Internet](<https://devfeed.tech/topics/internet.md>), [Web](<https://devfeed.tech/topics/web.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [Query (disambiguation)](<https://devfeed.tech/topics/query.md>), [cosine similarity search](<https://devfeed.tech/topics/cosine-similarity-search.md>)

Tags: [2021](<https://devfeed.tech/tags/2021.md>), [2022](<https://devfeed.tech/tags/2022.md>), [2023](<https://devfeed.tech/tags/2023.md>), [2025](<https://devfeed.tech/tags/2025.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [can](<https://devfeed.tech/tags/can.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [data](<https://devfeed.tech/tags/data.md>), [dataset](<https://devfeed.tech/tags/dataset.md>), [download](<https://devfeed.tech/tags/download.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [fast](<https://devfeed.tech/tags/fast.md>), [internet](<https://devfeed.tech/tags/internet.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [query](<https://devfeed.tech/tags/query.md>), [querying](<https://devfeed.tech/tags/querying.md>), [s3](<https://devfeed.tech/tags/s3.md>), [web](<https://devfeed.tech/tags/web.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

DuckDB can analyze Common Crawl's S3-hosted Parquet data without downloading the full dataset or using a cluster: sort-key statistics let it skip most row groups, while selecting only needed columns reduces reads. The article contrasts Common Crawl's broad, link-following crawl (which does not run JavaScript) with the HTTP Archive's technology-focused sampling and the Wayback Machine's URL history. Using quarterly web-graph hostname lists plus the columnar URL index, the author compares platform hostnames against established hosting references and scans 109.4 billion index rows to study GitHub Pages. The results suggest relative growth in hosted platforms and a 75% rise in crawled GitHub Pages user sites per million pages over four years, while projects per user site are roughly flat. These are proxies: Common Crawl counts pages it crawls, misses bot-blocked or unlinked sites, and crawl coverage varies, so the measurements do not represent all sites built.

### Source excerpt

Common Crawl publishes petabytes of web crawl data on S3. With DuckDB and MotherDuck you can query the Common Crawl dataset directly, no cluster and no download, and measure how fast the vibe-coded web is growing.

## Introducing prompt\_jev(): bringing Jev to Motherduck SQL

DevFeed: [Introducing prompt\_jev(): bringing Jev to Motherduck SQL](<https://devfeed.tech/articles/introducing-prompt-jev-bringing-jev-to-motherduck-sql-78942.md>)

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

Author: Hamilton Ulmer; Till Döhmen; Jacob Matson

Published: 2026-09-21T00:00:00Z

Content type: release

Language: en

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

Topics: [SQL](<https://devfeed.tech/topics/sql.md>), [prompt](<https://devfeed.tech/topics/prompt.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Machine learning](<https://devfeed.tech/topics/machine-learning.md>), [Enhancing AI model performance](<https://devfeed.tech/topics/enhancing-ai-model-performance.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2015](<https://devfeed.tech/tags/2015.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [administration](<https://devfeed.tech/tags/administration.md>), [aggregate](<https://devfeed.tech/tags/aggregate.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-model](<https://devfeed.tech/tags/ai-model.md>), [all](<https://devfeed.tech/tags/all.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [article](<https://devfeed.tech/tags/article.md>), [jev](<https://devfeed.tech/tags/jev.md>), [llm](<https://devfeed.tech/tags/llm.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [sql](<https://devfeed.tech/tags/sql.md>), [text](<https://devfeed.tech/tags/text.md>), [typesafe](<https://devfeed.tech/tags/typesafe.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

MotherDuck's `prompt_jev()` brings TypeSafe AI's Jev model into SQL as a function that takes text and returns a label, score, or yes/no decision with confidence, ready to filter, join, or aggregate in the same query. It is configured with a short prompt and choices, avoiding response parsing or the labeled examples and model maintenance needed for encoder workflows. The excerpt reports that on 100,000 AG News articles it matched frontier LLM accuracy in 40 seconds for $0.50, versus more than 30 minutes and $37 for a comparable LLM; tests at 1 million and 10 million rows reportedly showed similar or faster performance. These are vendor-reported benchmark results, not a general guarantee. `prompt_jev()` is available on paid MotherDuck plans.

### Source excerpt

Classify text in SQL with prompt_jev(), powered by TypeSafe's Jev. 100,000 rows labeled in 40 seconds for $0.50 at frontier-LLM accuracy. Live on paid plans.

## dbt Charts on MotherDuck: Declarative Dashboards

DevFeed: [dbt Charts on MotherDuck: Declarative Dashboards](<https://devfeed.tech/articles/dbt-charts-on-motherduck-declarative-dashboards-78935.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/dbt-Charts-on-MotherDuck-Declarative-Dashboards>)

Author: Alex Monahan

Published: 2026-09-21T00:00:00Z

Content type: article

Language: en

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

Topics: [Data visualization](<https://devfeed.tech/topics/data-visualization.md>), [dashboards](<https://devfeed.tech/topics/dashboards.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>), [JavaScript](<https://devfeed.tech/topics/javascript.md>), [User interface design](<https://devfeed.tech/topics/ui-design.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [apps](<https://devfeed.tech/tags/apps.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [browser](<https://devfeed.tech/tags/browser.md>), [build](<https://devfeed.tech/tags/build.md>), [charts](<https://devfeed.tech/tags/charts.md>), [claude](<https://devfeed.tech/tags/claude.md>), [command](<https://devfeed.tech/tags/command.md>), [compilation](<https://devfeed.tech/tags/compilation.md>), [config](<https://devfeed.tech/tags/config.md>), [dashboards](<https://devfeed.tech/tags/dashboards.md>), [dbt](<https://devfeed.tech/tags/dbt.md>), [declarative](<https://devfeed.tech/tags/declarative.md>), [live](<https://devfeed.tech/tags/live.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [package](<https://devfeed.tech/tags/package.md>), [pipeline](<https://devfeed.tech/tags/pipeline.md>), [refresh](<https://devfeed.tech/tags/refresh.md>), [yaml](<https://devfeed.tech/tags/yaml.md>)

### AI overview

The dbt package publishes declarative dbt Charts as MotherDuck Dives during the same `dbt build` that updates the data. Setup requires connecting the dbt project to MotherDuck, adding the package repository in `packages.yml`, and setting `save_charts_as_dives: true` in `dbt_project.yml`. At the build's `on-run-end` hook, SQL stores chart configurations and starts a serverless Python Flight; the Flight converts YAML to Vega Lite configurations and creates a Dive for each dashboard. Dives are React apps hosted by MotherDuck with direct data access, and the article says they refresh live when viewed. The package reuses dbt Charts' YAML-to-dashboard compilation, while MotherDuck's SQL, Python Flights, and TypeScript/React components handle data processing, orchestration, and visualization. The article describes Dives as capable of more than dashboards, but its example focuses on charts.

### Source excerpt

Define your dashboards in YAML with dbt Charts, then the same dbt build that runs your pipeline deploys them to MotherDuck Dives. Just 2 config changes! Your Dives will refresh live whenever your stakeholders take a look. See how we constructed the dbt package as well.

## DuckDB Ecosystem Newsletter : September 2026

DevFeed: [DuckDB Ecosystem Newsletter : September 2026](<https://devfeed.tech/articles/duckdb-ecosystem-newsletter-september-2026-78936.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/duckdb-ecosystem-newsletter-september-2026>)

Author: Simon Späti

Published: 2026-09-16T00:00:00Z

Content type: news

Language: en

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

Topics: [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [sql-server](<https://devfeed.tech/topics/sql-server.md>), [Apache Iceberg](<https://devfeed.tech/topics/apache-iceberg.md>), [etl](<https://devfeed.tech/topics/etl.md>), [data lake](<https://devfeed.tech/topics/data-lake.md>), [Amazon S3 Tables](<https://devfeed.tech/topics/amazon-s3-tables.md>), [streaming data analytics](<https://devfeed.tech/topics/streaming-data-analytics.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [alpha](<https://devfeed.tech/tags/alpha.md>), [aws](<https://devfeed.tech/tags/aws.md>), [canvas](<https://devfeed.tech/tags/canvas.md>), [community](<https://devfeed.tech/tags/community.md>), [data](<https://devfeed.tech/tags/data.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [ecosystem](<https://devfeed.tech/tags/ecosystem.md>), [functions](<https://devfeed.tech/tags/functions.md>), [infinite](<https://devfeed.tech/tags/infinite.md>), [java](<https://devfeed.tech/tags/java.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [newsletter](<https://devfeed.tech/tags/newsletter.md>), [plus](<https://devfeed.tech/tags/plus.md>), [september-2026](<https://devfeed.tech/tags/september-2026.md>), [server](<https://devfeed.tech/tags/server.md>), [sql](<https://devfeed.tech/tags/sql.md>), [sql-server](<https://devfeed.tech/tags/sql-server.md>), [table](<https://devfeed.tech/tags/table.md>), [tables](<https://devfeed.tech/tags/tables.md>)

### AI overview

The September 2026 newsletter reports two acquisitions and a DuckDB v2.0-alpha alongside practical ecosystem updates. DuckLabs joins AWS, while DuckDB, DuckLake, and Quack remain MIT-licensed under the DuckDB Foundation; MotherDuck acquired Tower.dev to integrate its runtime into MotherDuck Flights. The v2.0 preview includes Quack client/server mode, a new SQL parser and C API, and a new default storage format. Other examples include Java table functions that query JVM-accessible sources with predicate pushdown; Zarr stores queried as SQL tables; and the MSSQL extension's bulk-load improvements, including parallel writers and explicit string sizing. In a reported 38-million-row load, ingestion fell from 933 to 96 seconds. The excerpt also compares DuckLake and Iceberg in a CDC benchmark and describes Iceberg read/write support in DuckDB, with merge-on-read debt and table maintenance limitations.

### Source excerpt

The September 2026 DuckDB Ecosystem Newsletter: DuckLabs joins AWS and MotherDuck acquires Tower, DuckDB v2.0-alpha lands, table functions in pure Java, bulk loads into SQL Server, Zarr stores as SQL tables, an infinite canvas for your data, plus a community spotlight on Vladimir Gribanov.

## They All Write, Store, and Read Data: How to Pick The Right Database

DevFeed: [They All Write, Store, and Read Data: How to Pick The Right Database](<https://devfeed.tech/articles/they-all-write-store-and-read-data-how-to-pick-the-right-database-78938.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/how-to-pick-the-right-database>)

Author: Dumky de Wilde

Published: 2026-09-15T00:00:00Z

Content type: article

Language: en

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

Topics: [Database](<https://devfeed.tech/topics/database.md>), [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [sql-database](<https://devfeed.tech/topics/sql-database.md>), [graph-database](<https://devfeed.tech/topics/graph-database.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [compressing data with Parquet](<https://devfeed.tech/topics/compressing-data-with-parquet.md>), [database-consistency](<https://devfeed.tech/topics/database-consistency.md>), [LanceDB vector database](<https://devfeed.tech/topics/lancedb-vector-database.md>)

Tags: [aggregation](<https://devfeed.tech/tags/aggregation.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [analytics](<https://devfeed.tech/tags/analytics.md>), [atomic](<https://devfeed.tech/tags/atomic.md>), [audio](<https://devfeed.tech/tags/audio.md>), [cache](<https://devfeed.tech/tags/cache.md>), [data](<https://devfeed.tech/tags/data.md>), [database](<https://devfeed.tech/tags/database.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [postgres](<https://devfeed.tech/tags/postgres.md>), [questions](<https://devfeed.tech/tags/questions.md>), [sqlite](<https://devfeed.tech/tags/sqlite.md>), [store](<https://devfeed.tech/tags/store.md>), [write](<https://devfeed.tech/tags/write.md>)

### AI overview

Choose a database around the queries and workload you expect: SQLite suits small, mostly single-writer projects; Postgres is a strong general-purpose choice for shared transactional data; DuckDB and other analytical databases suit large aggregations. The article explains that DuckDB reduces work through columnar reads, row groups, predicate pushdown and pruning, compression, and query-plan optimization. Its embedded, in-process design can put compute alongside the application and avoid a shared database server. Other layouts fit specific needs: key/value stores provide fast known-key lookups but cannot answer unplanned queries; document databases allow flexible nested records but can make cross-record analysis costly; graph databases model relationships; vector indexes speed similarity search at the cost of approximate results; time-series databases optimize time-windowed data; and geospatial indexes support location queries. The guidance is workload-dependent: Postgres is presented as a practical default, with specialized systems useful when its performance or cost no longer fits.

### Source excerpt

SQLite, Postgres, DuckDB and 20 others: they all write, store and read data. The questions you ask every day tell you which one you need.

## Your DuckDB Flock Deserves A Mother: MotherDuck Now Offers Open Source DuckDB Support

DevFeed: [Your DuckDB Flock Deserves A Mother: MotherDuck Now Offers Open Source DuckDB Support](<https://devfeed.tech/articles/your-duckdb-flock-deserves-a-mother-motherduck-now-offers-open-source-duckdb-support-78937.md>)

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

Author: Kaitlyn Henry

Published: 2026-09-14T00:00:00Z

Content type: release

Language: en

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

Topics: [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [Support](<https://devfeed.tech/topics/support.md>)

Tags: [analytics](<https://devfeed.tech/tags/analytics.md>), [app](<https://devfeed.tech/tags/app.md>), [architect](<https://devfeed.tech/tags/architect.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [backend](<https://devfeed.tech/tags/backend.md>), [cloud](<https://devfeed.tech/tags/cloud.md>), [community](<https://devfeed.tech/tags/community.md>), [compatibility](<https://devfeed.tech/tags/compatibility.md>), [data](<https://devfeed.tech/tags/data.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [data-teams](<https://devfeed.tech/tags/data-teams.md>), [database](<https://devfeed.tech/tags/database.md>), [deployments](<https://devfeed.tech/tags/deployments.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [engine](<https://devfeed.tech/tags/engine.md>), [experts](<https://devfeed.tech/tags/experts.md>), [modern-data](<https://devfeed.tech/tags/modern-data.md>), [motherduck](<https://devfeed.tech/tags/motherduck.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [support](<https://devfeed.tech/tags/support.md>)

### AI overview

MotherDuck now sells commercial support for organizations running open-source DuckDB in production, including those unable to use hosted software because of regulatory or operational constraints. The offering includes SLA-backed response times, architecture guidance, workload optimization, upgrade compatibility checks, and help investigating failures, with access to MotherDuck database experts who may engage DuckDB's team when needed. The excerpt identifies data teams, analytics platforms, embedded and on-prem deployments, scaling organizations, and companies evaluating DuckDB as potential users. It does not specify pricing or the SLA terms.

### Source excerpt

We're launching commercial support for open-source DuckDB, so every organization can get the power of a modern data engine, with direct access to database experts.

## Why DuckDB 2.0 is faster

DevFeed: [Why DuckDB 2.0 is faster](<https://devfeed.tech/articles/why-duckdb-2-0-is-faster-78944.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/why-duckdb-20-is-faster>)

Author: Mehdi Ouazza

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

Content type: article

Language: en

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

Topics: [DuckDB](<https://devfeed.tech/topics/duckdb.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [JSON](<https://devfeed.tech/topics/json.md>), [Cache](<https://devfeed.tech/topics/cache.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [async](<https://devfeed.tech/tags/async.md>), [compatibility-mode](<https://devfeed.tech/tags/compatibility-mode.md>), [cte](<https://devfeed.tech/tags/cte.md>), [duckdb](<https://devfeed.tech/tags/duckdb.md>), [faster](<https://devfeed.tech/tags/faster.md>), [json](<https://devfeed.tech/tags/json.md>), [model](<https://devfeed.tech/tags/model.md>), [s3](<https://devfeed.tech/tags/s3.md>), [text](<https://devfeed.tech/tags/text.md>)

### AI overview

DuckDB 2.0 alpha improves remote reads, deep recursive queries, and JSON field access through separate changes. For S3 Parquet scans, async I/O downloads row groups ahead while workers decode them; it is enabled by default, with `read_ahead_depth = 0` restoring prior behavior. The author measured roughly 2-3x faster reads, but tiny files gain little because per-file round trips remain. Recursive CTEs now build a parent lookup once and expand from newly found rows, reducing repeated full-table scans; this chiefly benefits deep chains, while shallow hierarchies may see little improvement. VARIANT shreds consistently typed JSON fields into sub-columns and keeps irregular data in a queryable remainder. In the author's tests it used 2.7x less storage than JSON text and made queries on shredded fields about 6x faster, near typed columns; list casts were slower, and inconsistent value types prevent shredding. The excerpt also describes triggers with transition tables, nested schemas, DML in CTEs, and other additions. Results are from one M5 laptop and home internet, so the author advises running local benchmarks.

### Source excerpt

DuckDB 2.0 alpha vs 1.5.5 on one laptop: recursive CTEs up to 90x, VARIANT 6x faster than JSON text, async I/O 2.4x on S3. What changes, and how to model data to get the speedup.

## Agents Don't Query Like Humans Do

DevFeed: [Agents Don't Query Like Humans Do](<https://devfeed.tech/articles/agents-don-t-query-like-humans-do-78931.md>)

Original publisher: [Read original article](<https://motherduck.com/blog/agents-dont-query-like-humans-do>)

Author: Alex Monahan

Published: 2026-09-02T00:00:00Z

Content type: news

Language: en

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

Topics: [AI for SQL query generation](<https://devfeed.tech/topics/ai-for-sql-query-generation.md>), [scalable-architecture](<https://devfeed.tech/topics/scalable-architecture.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [self-service](<https://devfeed.tech/topics/self-service.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>)

Tags: [accelerated](<https://devfeed.tech/tags/accelerated.md>), [agentic-workloads](<https://devfeed.tech/tags/agentic-workloads.md>), [agents](<https://devfeed.tech/tags/agents.md>), [behavior](<https://devfeed.tech/tags/behavior.md>), [cache](<https://devfeed.tech/tags/cache.md>), [caching](<https://devfeed.tech/tags/caching.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [concurrent](<https://devfeed.tech/tags/concurrent.md>), [context](<https://devfeed.tech/tags/context.md>), [context-layer](<https://devfeed.tech/tags/context-layer.md>), [cost](<https://devfeed.tech/tags/cost.md>), [critical](<https://devfeed.tech/tags/critical.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [data-platform](<https://devfeed.tech/tags/data-platform.md>), [history](<https://devfeed.tech/tags/history.md>), [queries](<https://devfeed.tech/tags/queries.md>)

### AI overview

In a two-week comparison of MotherDuck UI queries (human) and MCP server queries (agent), agents issued far more queries, with the reported gap rising from 6x to 16x to 29x over three months; the text says agents now outnumber human users about 2 to 1 per organization. Agent query activity is burstier (median 4 seconds between queries versus 60 for humans) and typically smaller (median 7,000 rows scanned versus 200,000), including schema and metadata lookups. The article argues data platforms need to support high concurrency and rapid scale-up and scale-down, while keeping frequently reused data warm and making small exploratory queries fast and inexpensive. It also says a well-built context layer can improve query accuracy and reduce token use, citing 55% cost savings in the same benchmarks, and recommends local caching or catalog access to avoid repeated data pulls. The figures are based on a slice of query history and the stated UI/MCP segmentation.

### Source excerpt

We analyzed a slice of MotherDuck query history and found agents ran 29 times more queries than humans last month, a gap that has roughly doubled every month. The average organization now has twice as many agents as human users. Agents are the dominant consumers of the data platform, and our systems can't afford to stay the same.

[Next page](<https://devfeed.tech/sources/motherduck.md?cursor=WyIyMDI2LTA5LTAyVDAwOjAwOjAwKzAwOjAwIiwgIjhmNzBkYWRjLWZlMGQtNDM2OC1hODU3LTc5ZjM3YWM4YjZlNyJd>)