# Cockroach Labs

Welcome to this 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.

## How Does CockroachDB Automate SQL User Lifecycle Management?

DevFeed: [How Does CockroachDB Automate SQL User Lifecycle Management?](<https://devfeed.tech/articles/how-does-cockroachdb-automate-sql-user-lifecycle-management-23818.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/sql-user-lifecycle-management-automation>)

Author: Pritesh Lahoti,Biplav Saraf,Sourav Sarangi

Published: 2026-08-28T00:00:00Z

Content type: article

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [OpenID connect (OIDC)](<https://devfeed.tech/topics/oidc.md>), [active directory](<https://devfeed.tech/topics/active-directory.md>), [identity and access management](<https://devfeed.tech/topics/identity-and-access-management.md>), [Logging](<https://devfeed.tech/topics/logging.md>), [Single sign-on (SSO)](<https://devfeed.tech/topics/sso.md>), [Secure token management](<https://devfeed.tech/topics/secure-token-management.md>), [Entra ID](<https://devfeed.tech/topics/entra-id.md>), [okta](<https://devfeed.tech/topics/okta.md>), [Cockroach Labs](<https://devfeed.tech/topics/cockroach-labs.md>)

Tags: [access-management](<https://devfeed.tech/tags/access-management.md>), [active-directory](<https://devfeed.tech/tags/active-directory.md>), [automate](<https://devfeed.tech/tags/automate.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [directory](<https://devfeed.tech/tags/directory.md>), [enterprises](<https://devfeed.tech/tags/enterprises.md>), [entra-id](<https://devfeed.tech/tags/entra-id.md>), [iam](<https://devfeed.tech/tags/iam.md>), [identity](<https://devfeed.tech/tags/identity.md>), [identity-and-access-management](<https://devfeed.tech/tags/identity-and-access-management.md>), [identity-provider](<https://devfeed.tech/tags/identity-provider.md>), [idp](<https://devfeed.tech/tags/idp.md>), [lifecycle](<https://devfeed.tech/tags/lifecycle.md>), [management](<https://devfeed.tech/tags/management.md>), [microsoft](<https://devfeed.tech/tags/microsoft.md>), [microsoft-entra-id](<https://devfeed.tech/tags/microsoft-entra-id.md>), [okta](<https://devfeed.tech/tags/okta.md>), [ory](<https://devfeed.tech/tags/ory.md>), [sql](<https://devfeed.tech/tags/sql.md>), [user](<https://devfeed.tech/tags/user.md>)

### AI overview

CockroachDB automates SQL user lifecycle management for externally authenticated users: LDAP, JWT, or OIDC can create a SQL user on first successful login and stamp its immutable PROVISIONSRC with the authentication method and identity provider. When authorization is enabled, each login replaces the user's synchronized role memberships with current IdP groups; matching database roles must already exist, and changes take effect at the next login. Operators can filter users by source and estimated last login to review accounts, then use DROP PROVISIONED ROLES for scoped bulk cleanup. It requires CREATEROLE, protects root and admin, skips users with dependencies with a NOTICE, and records drop audit events. Login time is best-effort, not a complete authentication log; users with no recorded login are not included in last-login-before queries and need separate review. The article says PostgreSQL drivers can carry LDAP passwords or JWTs through the standard password field; OIDC is for DB Console browser login, with JWTs available for subsequent SQL connections.

### Source excerpt

Fortune 1000 enterprises widely rely on major Identity Provider (IdP) and Identity and Access Management (IAM) platforms like Okta, Microsoft Entra ID, Microsoft Active Directory, and Ory.

## Core Banking Modernization with Temenos Core and CockroachDB

DevFeed: [Core Banking Modernization with Temenos Core and CockroachDB](<https://devfeed.tech/articles/core-banking-modernization-with-temenos-core-and-cockroachdb-23772.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/core-banking-modernization-temenos-cockroachdb>)

Author: Nanda Badrappan,Muruga Balakrishnan

Published: 2026-08-24T00:00:00Z

Content type: article

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [Legacy Modernization](<https://devfeed.tech/topics/legacy-modernization.md>), [it modernization](<https://devfeed.tech/topics/it-modernization.md>), [Cloud Architecture](<https://devfeed.tech/topics/cloud-architecture.md>), [legacy systems](<https://devfeed.tech/topics/legacy-systems.md>)

Tags: [banking](<https://devfeed.tech/tags/banking.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [core-banking](<https://devfeed.tech/tags/core-banking.md>), [financial-services](<https://devfeed.tech/tags/financial-services.md>), [modernization](<https://devfeed.tech/tags/modernization.md>)

### AI overview

Joint testing of Temenos Core on CockroachDB v25.4.2 covered 2 million transactions across 20,000 accounts, 20 currencies, high-volume processing, and Close of Business workloads in on-premises Kubernetes and Azure environments. The partners report near-linear scaling in tested configurations as resources were added, and that HVT Merge Service and day-end processing ran alongside online workloads without reported degradation. CockroachDB supplies distributed SQL, serializable transactions, Raft-based replication and automatic recovery; Temenos Core supplies banking application functions. Integration uses TAFJ's PostgreSQL-flavored JDBC configuration and required targeted transaction-isolation, zone-configuration, and messaging optimizations documented in a reference architecture. Detailed benchmark results are available through the joint engagement team. Results are environment-specific and may vary; larger-scale testing and migration tooling are planned.

### Source excerpt

Banking has become an always-on business. Customers expect instant payments, accurate balances, and continuous access to financial services...

## A2A Is Now an Open Standard. The Data Layer Underneath It Isn't.

DevFeed: [A2A Is Now an Open Standard. The Data Layer Underneath It Isn't.](<https://devfeed.tech/articles/a2a-is-now-an-open-standard-the-data-layer-underneath-it-isn-t-23729.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/a2a-agent-state-data-layer>)

Author: Quentin Packard

Published: 2026-08-19T00:00:00Z

Content type: article

Language: en

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

Topics: [A2A protocol](<https://devfeed.tech/topics/a2a-protocol.md>), [agent to agent communication](<https://devfeed.tech/topics/agent-to-agent-communication.md>), [Distributed Systems & Parallel Computing](<https://devfeed.tech/topics/distributed-systems-parallel-computing.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [linux foundation](<https://devfeed.tech/topics/linux-foundation.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2025](<https://devfeed.tech/tags/2025.md>), [a2a](<https://devfeed.tech/tags/a2a.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-communication](<https://devfeed.tech/tags/agent-communication.md>), [agent-to-agent-communication](<https://devfeed.tech/tags/agent-to-agent-communication.md>), [communication](<https://devfeed.tech/tags/communication.md>), [data](<https://devfeed.tech/tags/data.md>), [foundation](<https://devfeed.tech/tags/foundation.md>), [google](<https://devfeed.tech/tags/google.md>), [isn-t](<https://devfeed.tech/tags/isn-t.md>), [it](<https://devfeed.tech/tags/it.md>), [layer](<https://devfeed.tech/tags/layer.md>), [linux](<https://devfeed.tech/tags/linux.md>), [linux-foundation](<https://devfeed.tech/tags/linux-foundation.md>), [open](<https://devfeed.tech/tags/open.md>), [open-standard](<https://devfeed.tech/tags/open-standard.md>), [protocol](<https://devfeed.tech/tags/protocol.md>), [standard](<https://devfeed.tech/tags/standard.md>), [the-linux-foundation](<https://devfeed.tech/tags/the-linux-foundation.md>)

### AI overview

A2A standardizes how agents delegate tasks and return results, but leaves shared state, memory, identity, consistency, and cross-organization auditing to system designers. MCP connects an agent to tools and data, while AGENTS.md supplies repository instructions; neither fills that state-layer gap. The excerpt argues that when independent agents cross organizational boundaries and write state, teams need a governed, queryable context store, a defined consistency model, durable checkpoints, scoped delegation credentials with expiry, and agreed cross-boundary logs. It describes serializable isolation as one way to detect conflicting writes, while noting conditional writes or row locks can prevent a specific race. Workflow engines can resume execution but do not govern shared data outside their boundary. These measures may be unnecessary for agents within one runtime, tool-only setups, read-only specialists, or disposable workflows.

### Source excerpt

In April 2025, Google released a protocol for agent-to-agent communication. Within three months, Google had donated it to the Linux Foundation...

## Multi-Tenant AI Agents: Why Data Isolation Starts at the Database

DevFeed: [Multi-Tenant AI Agents: Why Data Isolation Starts at the Database](<https://devfeed.tech/articles/multi-tenant-ai-agents-why-data-isolation-starts-at-the-database-23798.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/multi-tenant-ai-agents-why-data-isolation-starts-at-the-database>)

Author: Quentin Packard

Published: 2026-08-11T00:00:00Z

Content type: article

Language: en

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

Topics: [tenant data protection](<https://devfeed.tech/topics/tenant-data-protection.md>), [ai security](<https://devfeed.tech/topics/ai-security.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [rdbms](<https://devfeed.tech/topics/rdbms.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-security](<https://devfeed.tech/tags/agent-security.md>), [agentic](<https://devfeed.tech/tags/agentic.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-security](<https://devfeed.tech/tags/ai-agent-security.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [also](<https://devfeed.tech/tags/also.md>), [api](<https://devfeed.tech/tags/api.md>), [but](<https://devfeed.tech/tags/but.md>), [data](<https://devfeed.tech/tags/data.md>), [database](<https://devfeed.tech/tags/database.md>), [effective](<https://devfeed.tech/tags/effective.md>), [features](<https://devfeed.tech/tags/features.md>), [focus](<https://devfeed.tech/tags/focus.md>), [isolation](<https://devfeed.tech/tags/isolation.md>), [layer](<https://devfeed.tech/tags/layer.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [saas](<https://devfeed.tech/tags/saas.md>)

### AI overview

The provided text argues that multi-tenant AI agents need tenant boundaries enforced in the database because an agent can act on improperly retrieved data and amplify a breach through tool calls. It describes CockroachDB row-level security (RLS) for shared tables: policies apply to each query, while the application must set tenant context on every connection. App filters can remain for performance and defense in depth. CockroachDB-specific REGIONAL BY ROW placement can pin tenant data to a home region for residency needs, without separate regional clusters. The text also notes RLS does not stop malicious instructions stored in a tenant's own data, inference-cache timing leaks, unpartitioned memory or conversation history, or overly broad service-account permissions; these require additional controls. It says RLS may add maintenance overhead without security gain for single-tenant deployments or tenants already isolated in separate databases or clusters.

### Source excerpt

Most SaaS teams shipping agentic features focus on prompt safety and API-layer filtering. But effective AI agent security also depends on...

## DORA Compliance for AI Agents: Database Requirements Before Deployment

DevFeed: [DORA Compliance for AI Agents: Database Requirements Before Deployment](<https://devfeed.tech/articles/dora-compliance-for-ai-agents-database-requirements-before-deployment-23781.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/dora-database-requirements-ai-agents>)

Author: Quentin Packard

Published: 2026-08-07T00:00:00Z

Content type: article

Language: en

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

Topics: [Database](<https://devfeed.tech/topics/database.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Resilience](<https://devfeed.tech/topics/resilience.md>), [audit](<https://devfeed.tech/topics/audit.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [integrity](<https://devfeed.tech/topics/integrity.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [audit](<https://devfeed.tech/tags/audit.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [database](<https://devfeed.tech/tags/database.md>), [eu](<https://devfeed.tech/tags/eu.md>), [gdpr](<https://devfeed.tech/tags/gdpr.md>), [integrity](<https://devfeed.tech/tags/integrity.md>), [resilience](<https://devfeed.tech/tags/resilience.md>), [transactions](<https://devfeed.tech/tags/transactions.md>)

### AI overview

This article explains database capabilities needed by financial AI agents operating under DORA, the EU AI Act, and GDPR. It focuses on auditability, agent identity attribution, strong read consistency, idempotency, rollback and compensating transactions, and data residency. It argues that these database properties help organizations demonstrate operational resilience, traceability, and transaction integrity before deployment.

### Source excerpt

Financial AI agents operating under DORA, the EU AI Act, and GDPR need infrastructure that supports operational resilience, traceability, and reliable transaction processing.

## The Database at 550 Kilometers: What Orbital Computing Means for Distributed Databases

DevFeed: [The Database at 550 Kilometers: What Orbital Computing Means for Distributed Databases](<https://devfeed.tech/articles/the-database-at-550-kilometers-what-orbital-computing-means-for-distributed-databases-23801.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/orbital-computing-distributed-databases>)

Author: Isaac Wong

Published: 2026-08-06T00:00:00Z

Content type: opinion

Language: en

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

Topics: [data centers](<https://devfeed.tech/topics/data-centers.md>), [Database](<https://devfeed.tech/topics/database.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [GPU](<https://devfeed.tech/topics/gpu.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [compute](<https://devfeed.tech/tags/compute.md>), [data-center](<https://devfeed.tech/tags/data-center.md>), [data-centers](<https://devfeed.tech/tags/data-centers.md>), [database](<https://devfeed.tech/tags/database.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [energy](<https://devfeed.tech/tags/energy.md>), [gpu](<https://devfeed.tech/tags/gpu.md>), [inference](<https://devfeed.tech/tags/inference.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [nvidia](<https://devfeed.tech/tags/nvidia.md>), [solar](<https://devfeed.tech/tags/solar.md>), [space](<https://devfeed.tech/tags/space.md>), [spacex](<https://devfeed.tech/tags/spacex.md>)

### AI overview

The article examines orbital data centers as a possible response to the energy, cooling, and land constraints of terrestrial facilities. It describes proposed satellite-based computing projects, including an orbital Nvidia H100 Gemini inference workload, and outlines the roles of solar power, radiative cooling, and open orbit.

### Source excerpt

In Ashburn, Virginia, a row of servers draws 40 megawatts from the grid and exhales it as heat.

## CockroachDB Kubernetes Operator launches with full lifecycle automation

DevFeed: [CockroachDB Kubernetes Operator launches with full lifecycle automation](<https://devfeed.tech/articles/database-aware-kubernetes-operations-are-here-23765.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/cockroachdb-kubernetes-operator>)

Author: Ethan Shmargad,Pritesh Lahoti

Published: 2026-08-05T00:00:00Z

Content type: release

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [Kubernetes](<https://devfeed.tech/topics/kubernetes.md>), [Deployment](<https://devfeed.tech/topics/deployment.md>), [Automation](<https://devfeed.tech/topics/automation.md>), [migration](<https://devfeed.tech/topics/migration.md>), [Scalability](<https://devfeed.tech/topics/scalability.md>)

Tags: [automation](<https://devfeed.tech/tags/automation.md>), [cockroach-labs](<https://devfeed.tech/tags/cockroach-labs.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [database](<https://devfeed.tech/tags/database.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [kubernetes](<https://devfeed.tech/tags/kubernetes.md>), [migration](<https://devfeed.tech/tags/migration.md>), [production](<https://devfeed.tech/tags/production.md>), [scalability](<https://devfeed.tech/tags/scalability.md>)

### AI overview

Cockroach Labs announces the CockroachDB Kubernetes Operator, designed to connect CockroachDB-aware operations with Kubernetes. It provides lifecycle automation for deployment, upgrades, scaling, certificate rotation, and storage management, and supports multi-region deployments and workflows including Helm, GitOps, private registries, and air-gapped environments.

### Source excerpt

The production-proven K8s operator is now available with full lifecycle automation, zero-downtime migration, as well as the flexibility and scalability to fit how your team actually works.

## How to Consolidate Your Database Stack for Production AI

DevFeed: [How to Consolidate Your Database Stack for Production AI](<https://devfeed.tech/articles/how-to-consolidate-your-database-stack-for-production-ai-23773.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/database-consolidation-production-ai>)

Author: David Weiss

Published: 2026-07-28T00:00:00Z

Content type: tutorial

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [Database](<https://devfeed.tech/topics/database.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [databases](<https://devfeed.tech/tags/databases.md>), [guide](<https://devfeed.tech/tags/guide.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [latency](<https://devfeed.tech/tags/latency.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [redis](<https://devfeed.tech/tags/redis.md>), [resilience](<https://devfeed.tech/tags/resilience.md>), [scale](<https://devfeed.tech/tags/scale.md>), [vector-database](<https://devfeed.tech/tags/vector-database.md>), [vector-search](<https://devfeed.tech/tags/vector-search.md>)

### AI overview

This guide examines consolidating PostgreSQL, Redis, and a dedicated vector database into CockroachDB for production AI applications. It discusses operational complexity, consistency, latency, resilience, scale, and the tradeoffs of using a single operational database.

### Source excerpt

When builders ship AI-powered applications, the data layer quietly becomes the hardest part of the stack.

## Why Your AI Agent Has an Identity Problem

DevFeed: [Why Your AI Agent Has an Identity Problem](<https://devfeed.tech/articles/why-your-ai-agent-has-an-identity-problem-23743.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/ai-agent-identity-security>)

Author: Quentin Packard

Published: 2026-07-17T00:00:00Z

Content type: article

Language: en

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

Topics: [identity and access management](<https://devfeed.tech/topics/identity-and-access-management.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [least privilege](<https://devfeed.tech/topics/least-privilege.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [access-controls](<https://devfeed.tech/tags/access-controls.md>), [access-management](<https://devfeed.tech/tags/access-management.md>), [account](<https://devfeed.tech/tags/account.md>), [accountability](<https://devfeed.tech/tags/accountability.md>), [agent](<https://devfeed.tech/tags/agent.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [apis](<https://devfeed.tech/tags/apis.md>), [call](<https://devfeed.tech/tags/call.md>), [customer](<https://devfeed.tech/tags/customer.md>), [external](<https://devfeed.tech/tags/external.md>), [identity](<https://devfeed.tech/tags/identity.md>), [just](<https://devfeed.tech/tags/just.md>), [records](<https://devfeed.tech/tags/records.md>), [team](<https://devfeed.tech/tags/team.md>), [transactions](<https://devfeed.tech/tags/transactions.md>), [users](<https://devfeed.tech/tags/users.md>)

### AI overview

The article argues that shared, broadly privileged service accounts leave AI agents hard to distinguish and their actions difficult to audit, creating security and compliance risks. It describes four production pitfalls: shared credentials, secrets exposed in model context, privilege inheritance across delegation, and human-focused audit logs that omit agent context. It recommends per-session, per-agent short-lived scoped credentials held by the tool-execution layer; least-privilege roles enforced at the database, including RBAC and row-level security; explicit delegation records with permissions narrowed at each handoff; and structured execution records linking sessions, users, model versions, inputs, tool calls, delegation, data access, costs, and outputs. The article notes that detailed row-level security and immutable-log designs are deferred to later articles.

### Source excerpt

The agent your team just shipped to external users has the ability to read customer records, execute transactions, and call external APIs.

## How BetterTracker Replaced Its Vector Store with CockroachDB

DevFeed: [How BetterTracker Replaced Its Vector Store with CockroachDB](<https://devfeed.tech/articles/how-bettertracker-replaced-its-vector-store-with-cockroachdb-23754.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/bettertracker-replaced-vector-store-cockroachdb>)

Author: Yohan Shirazi

Published: 2026-07-15T00:00:00Z

Content type: news

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>), [Database](<https://devfeed.tech/topics/database.md>), [vector embeddings in AI](<https://devfeed.tech/topics/vector-embeddings-in-ai.md>), [database-scalability](<https://devfeed.tech/topics/database-scalability.md>), [observability](<https://devfeed.tech/topics/observability.md>), [Data pipelines](<https://devfeed.tech/topics/data-pipelines.md>), [sql-database](<https://devfeed.tech/topics/sql-database.md>), [Vectors & LLMs](<https://devfeed.tech/topics/vectors-llms.md>), [data-engineering](<https://devfeed.tech/topics/data-engineering.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [implementation](<https://devfeed.tech/topics/implementation.md>), [audit](<https://devfeed.tech/topics/audit.md>)

Tags: [2025](<https://devfeed.tech/tags/2025.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [ai-assistant](<https://devfeed.tech/tags/ai-assistant.md>), [ai-features](<https://devfeed.tech/tags/ai-features.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-platform](<https://devfeed.tech/tags/ai-platform.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [costs](<https://devfeed.tech/tags/costs.md>), [database](<https://devfeed.tech/tags/database.md>), [risk](<https://devfeed.tech/tags/risk.md>), [search](<https://devfeed.tech/tags/search.md>), [vector](<https://devfeed.tech/tags/vector.md>), [vector-database](<https://devfeed.tech/tags/vector-database.md>), [vector-search](<https://devfeed.tech/tags/vector-search.md>)

### AI overview

BetterTracker replaced its standalone vector store by moving embeddings and production vector search into the CockroachDB cluster already running its transactional platform. The separate store had added API latency, ETL synchronization, debugging overhead, baseline and per-index or per-agent costs, and a second compliance boundary. In a structured proof of concept, the team used CockroachDB vector indexes, primary-key-scoped indexes for account isolation, and quantization that compressed embeddings from about 3 KB to about 200 bytes. They tested on the free Serverless tier, migrated company records and a vendor catalog, redirected AWS Lambda functions, then shut down the separate store. The source reports production cosine and L2 searches for tool classification and company deduplication, no measurable regression to transactional workloads, and a cutover without interruption. It says vector queries are billed as standard transactions and the existing SOC2, PCI, and HIPAA posture remained within CockroachDB Cloud Advanced. Specific time savings are not quantified; expansion to more tables and embeddings is a roadmap.

### Source excerpt

BetterTracker eliminated a standalone vector database by running OLTP and vector search together on CockroachDB--cutting costs, complexity, and compliance risk in one move.

## Announcing the CIS Benchmark for CockroachDB v25.x

DevFeed: [Announcing the CIS Benchmark for CockroachDB v25.x](<https://devfeed.tech/articles/announcing-the-cis-benchmark-for-cockroachdb-v25-x-23757.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/cis-benchmark-cockroachdb-security>)

Author: Adam Brennick,Ayog Mohanty

Published: 2026-07-14T00:00:00Z

Content type: release

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [Cryptography](<https://devfeed.tech/topics/cryptography.md>), [observability](<https://devfeed.tech/topics/observability.md>)

Tags: [audit](<https://devfeed.tech/tags/audit.md>), [auditing](<https://devfeed.tech/tags/auditing.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [aws](<https://devfeed.tech/tags/aws.md>), [azure](<https://devfeed.tech/tags/azure.md>), [backup](<https://devfeed.tech/tags/backup.md>), [backups](<https://devfeed.tech/tags/backups.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [best-practices](<https://devfeed.tech/tags/best-practices.md>), [cluster](<https://devfeed.tech/tags/cluster.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [cockroach-labs](<https://devfeed.tech/tags/cockroach-labs.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [collaboration](<https://devfeed.tech/tags/collaboration.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [configuration-drift](<https://devfeed.tech/tags/configuration-drift.md>), [consensus](<https://devfeed.tech/tags/consensus.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

CIS published version 1.0.0 of its CockroachDB v25.x Benchmark, a consensus-developed security baseline for self-hosted deployments. It contains 30 Level 1 recommendations across six areas: installation and patches, system hardening and topology, logging and monitoring, access and authorization, data protection, and database settings. Controls include TLS mutual authentication, prohibiting --insecure in production, least-privilege accounts, secure logging, hardened authentication, encrypted backups, and tested restores. The free PDF provides audit steps, remediation guidance, and CIS Controls mappings. An open-source auditing skill is also described as automating assessment of all 30 controls. The benchmark is intended to support consistent hardening and security reviews; the text notes that encryption at rest does not protect backup files.

### Source excerpt

We're proud to announce that the Center for Internet Security (CIS) has published the CIS CockroachDB v25.x Benchmark.

## Durable Execution with DBOS and CockroachDB

DevFeed: [Durable Execution with DBOS and CockroachDB](<https://devfeed.tech/articles/durable-execution-with-dbos-and-cockroachdb-23782.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/embedded-durable-execution-dbos-cockroachdb>)

Author: Amine El Kouhen, Ph.D.

Published: 2026-07-08T00:00:00Z

Content type: article

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [Persistence](<https://devfeed.tech/topics/persistence.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [long-running](<https://devfeed.tech/topics/long-running.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [gRPC](<https://devfeed.tech/topics/grpc.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [3](<https://devfeed.tech/tags/3.md>), [actions](<https://devfeed.tech/tags/actions.md>), [active-active](<https://devfeed.tech/tags/active-active.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [applications](<https://devfeed.tech/tags/applications.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [durable-execution](<https://devfeed.tech/tags/durable-execution.md>), [execution](<https://devfeed.tech/tags/execution.md>), [inference](<https://devfeed.tech/tags/inference.md>), [long-running](<https://devfeed.tech/tags/long-running.md>), [long-running-agents](<https://devfeed.tech/tags/long-running-agents.md>), [modern](<https://devfeed.tech/tags/modern.md>), [plan](<https://devfeed.tech/tags/plan.md>), [retry](<https://devfeed.tech/tags/retry.md>), [running](<https://devfeed.tech/tags/running.md>), [time](<https://devfeed.tech/tags/time.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

DBOS embeds durable workflow execution in a Python or TypeScript application and checkpoints workflow inputs, completed step outputs, and final status to a PostgreSQL-compatible database. After a crash, it replays the workflow from its original inputs and skips checkpointed steps; safe recovery requires deterministic workflow bodies and idempotent steps. With CockroachDB, LISTEN/NOTIFY must be disabled, so DBOS polls instead. The excerpt reports PostgreSQL faster for raw writes and at low concurrency, while both databases reach about 117 workflows per second in the tested workload; it attributes this limit to DBOS's sequential acknowledged commits. It describes CockroachDB's multi-node, zone-redundant architecture as a path to scaling beyond PostgreSQL's measured single-node ceiling, while presenting projected higher-node throughput rather than benchmarked results.

### Source excerpt

Modern AI applications are no longer single-shot inference calls. They are long-running agents that plan, act, observe, and retry across time.

## How CockroachDB and IBM LinuxONE Rockhopper 5 Power Resilient AI Infrastructure

DevFeed: [How CockroachDB and IBM LinuxONE Rockhopper 5 Power Resilient AI Infrastructure](<https://devfeed.tech/articles/how-cockroachdb-and-ibm-linuxone-rockhopper-5-power-resilient-ai-infrastructure-23791.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/ibm-linuxone-rockhopper-5-ai-infrastructure>)

Author: Kyle Basile

Published: 2026-07-07T00:00:00Z

Content type: article

Language: en

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

Topics: [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [NVIDIA DGX](<https://devfeed.tech/topics/nvidia-dgx.md>), [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [ibm](<https://devfeed.tech/topics/ibm.md>), [Database](<https://devfeed.tech/topics/database.md>), [AI Development](<https://devfeed.tech/topics/ai-development.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [availability](<https://devfeed.tech/tags/availability.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [customers](<https://devfeed.tech/tags/customers.md>), [data](<https://devfeed.tech/tags/data.md>), [digital](<https://devfeed.tech/tags/digital.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [failover](<https://devfeed.tech/tags/failover.md>), [financial](<https://devfeed.tech/tags/financial.md>), [ibm](<https://devfeed.tech/tags/ibm.md>), [identity](<https://devfeed.tech/tags/identity.md>), [identity-and-access](<https://devfeed.tech/tags/identity-and-access.md>), [identity-and-access-management](<https://devfeed.tech/tags/identity-and-access-management.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

The article explains how enterprise AI workloads increase requirements for continuous data availability, high concurrency, real-time decision-making, and elastic scaling. It presents IBM LinuxONE Rockhopper 5 and CockroachDB as infrastructure and data-platform components for resilient AI applications, including semantic retrieval, embeddings, agent memory, and transactional consistency.

### Source excerpt

Why does AI require a new approach to infrastructure?

## RoachFest London 2026 Recap: An Astronaut, AI Agents, Amazingness

DevFeed: [RoachFest London 2026 Recap: An Astronaut, AI Agents, Amazingness](<https://devfeed.tech/articles/roachfest-london-2026-recap-an-astronaut-ai-agents-amazingness-23809.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/roachfest-london-2026-recap>)

Author: Rob Reid

Published: 2026-07-02T00:00:00Z

Content type: article

Language: en

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

Topics: [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [hierarchical agent systems](<https://devfeed.tech/topics/hierarchical-agent-systems.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [long-running](<https://devfeed.tech/topics/long-running.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [astronaut](<https://devfeed.tech/tags/astronaut.md>), [london](<https://devfeed.tech/tags/london.md>), [recap](<https://devfeed.tech/tags/recap.md>)

### AI overview

RoachFest London 2026 showcased Cockroach Labs' vision for database infrastructure shaped by agentic AI: a purpose-built storage layer claimed to deliver 10x throughput and 70% lower storage costs, compute separated from storage, millions of virtual clusters per physical cluster, and specialized operations agents working under human-defined policy. The roadmap also included an AI migration assistant targeting a 50% reduction in PostgreSQL and MySQL migration time and cost, BYOC generally available on AWS, GCP, and Azure, and a Kubernetes Operator planned for general availability in July. Customer sessions described Form3's active/active/active payment platform across three cloud providers and Primer's zero-downtime migration from PostgreSQL on RDS using different strategies for different workloads. The excerpt also covers discussions of agent memory tradeoffs and astronaut Tim Peake's lessons on resilience. These are event-recap claims and roadmap statements, not independent performance evaluations.

### Source excerpt

RoachFest London 2026 has been and gone, and I'm still buzzing.

## Why Agent Loops Fail in Production (and the Database Patterns That Fix Them)

DevFeed: [Why Agent Loops Fail in Production (and the Database Patterns That Fix Them)](<https://devfeed.tech/articles/why-agent-loops-fail-in-production-and-the-database-patterns-that-fix-them-23732.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/agent-loops-production-database-patterns>)

Author: Quentin Packard

Published: 2026-07-01T00:00:00Z

Content type: article

Language: en

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

Topics: [ReAct agent Python](<https://devfeed.tech/topics/react-agent-python.md>), [Database](<https://devfeed.tech/topics/database.md>), [database-consistency](<https://devfeed.tech/topics/database-consistency.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [software-architecture](<https://devfeed.tech/topics/software-architecture.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [accountability](<https://devfeed.tech/tags/accountability.md>), [actions](<https://devfeed.tech/tags/actions.md>), [added](<https://devfeed.tech/tags/added.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-reliability](<https://devfeed.tech/tags/agent-reliability.md>), [agent-state](<https://devfeed.tech/tags/agent-state.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [database](<https://devfeed.tech/tags/database.md>), [fail](<https://devfeed.tech/tags/fail.md>), [little](<https://devfeed.tech/tags/little.md>), [loops](<https://devfeed.tech/tags/loops.md>), [model](<https://devfeed.tech/tags/model.md>), [patterns](<https://devfeed.tech/tags/patterns.md>), [production](<https://devfeed.tech/tags/production.md>), [state](<https://devfeed.tech/tags/state.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

Production agent loops can fail even when the model reasons correctly because concurrent writes, retries, crashes, stale memory and lost approval state leave inconsistent or unrecoverable workflow data. The article maps seven database-related failure modes to patterns including transactions for related writes, narrowly scoped roles, durable checkpoint and approval tables, and append-only, tamper-evident audit records tied to actions and credentials. It also describes temporal reads in CockroachDB for inspecting past state without restoring the live database. Examples target PostgreSQL 16; CockroachDB-specific syntax uses v24.3, and the article notes limits on combining historical and live reads in one statement. The supplied text presents these as practical patterns and code examples, while emphasizing that durable recovery must also account for workflow position, memory and downstream side effects.

### Source excerpt

Agent loops fail in production for reasons that have little to do with the model, and everything to do with what happens to their state between iterations.

## PostgreSQL-Compatible Databases for AI at Scale: What to Evaluate from Day One

DevFeed: [PostgreSQL-Compatible Databases for AI at Scale: What to Evaluate from Day One](<https://devfeed.tech/articles/postgresql-compatible-databases-for-ai-at-scale-what-to-evaluate-from-day-one-23802.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/postgresql-compatible-databases-ai-scale>)

Author: David Weiss

Published: 2026-06-29T00:00:00Z

Content type: article

Language: en

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

Topics: [PostgreSQL](<https://devfeed.tech/topics/postgresql.md>), [Database](<https://devfeed.tech/topics/database.md>), [SQL](<https://devfeed.tech/topics/sql.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-memory](<https://devfeed.tech/tags/agent-memory.md>), [agent-state](<https://devfeed.tech/tags/agent-state.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-applications](<https://devfeed.tech/tags/ai-applications.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [database](<https://devfeed.tech/tags/database.md>), [databases](<https://devfeed.tech/tags/databases.md>), [postgresql](<https://devfeed.tech/tags/postgresql.md>), [postgresql-compatible](<https://devfeed.tech/tags/postgresql-compatible.md>), [scale](<https://devfeed.tech/tags/scale.md>)

### AI overview

The article argues that AI workloads can expose single-node PostgreSQL limits early: agent fan-out and concurrency may exhaust connections, concentrate writes, and increase failure risk, while separating vector search from transactional data creates synchronization and duplication work. It recommends evaluating PostgreSQL-compatible databases for horizontal read and write scaling, distributed ACID transactions, vector search alongside operational data, multi-region deployment, zero-downtime schema changes, and PostgreSQL compatibility. It contrasts distributed SQL, which it says can scale while preserving transactional consistency, with managed single-region PostgreSQL, which retains a single-node ceiling, and NoSQL, which offers horizontal scale but sacrifices transactional consistency. CockroachDB is presented as an example, with automatic data distribution, serializable isolation, C-SPANN vector indexing, and a managed MCP Server. The excerpt presents these as the article's recommendations and product claims.

### Source excerpt

The database you choose at the start of an AI project is the one you'll be living with, or paying to escape, for years.

## The Thundering Herd Problem in Agentic AI: Why Traditional Fixes Fall Short

DevFeed: [The Thundering Herd Problem in Agentic AI: Why Traditional Fixes Fall Short](<https://devfeed.tech/articles/the-thundering-herd-problem-in-agentic-ai-why-traditional-fixes-fall-short-23739.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/agentic-ai-thundering-herd-problem>)

Author: Quentin Packard

Published: 2026-06-19T00:00:00Z

Content type: article

Language: en

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

Topics: [performance-engineering](<https://devfeed.tech/topics/performance-engineering.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [synchronization](<https://devfeed.tech/topics/synchronization.md>), [site-reliability-engineering](<https://devfeed.tech/topics/site-reliability-engineering.md>), [ai observability](<https://devfeed.tech/topics/ai-observability.md>), [observability](<https://devfeed.tech/topics/observability.md>), [LLM observability](<https://devfeed.tech/topics/llm-observability.md>), [Post Mortem](<https://devfeed.tech/topics/post-mortem.md>), [AI for customer service automation](<https://devfeed.tech/topics/ai-for-customer-service-automation.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2020](<https://devfeed.tech/tags/2020.md>), [2025](<https://devfeed.tech/tags/2025.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [act](<https://devfeed.tech/tags/act.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agent-framework](<https://devfeed.tech/tags/agent-framework.md>), [agent-orchestration](<https://devfeed.tech/tags/agent-orchestration.md>), [agent-state](<https://devfeed.tech/tags/agent-state.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-ai-architecture](<https://devfeed.tech/tags/agentic-ai-architecture.md>), [agentic-ai-security](<https://devfeed.tech/tags/agentic-ai-security.md>), [ai](<https://devfeed.tech/tags/ai.md>), [fall](<https://devfeed.tech/tags/fall.md>), [herd](<https://devfeed.tech/tags/herd.md>), [thundering-herd](<https://devfeed.tech/tags/thundering-herd.md>)

### AI overview

Agentic AI can create thundering-herd load internally: parallel agents and tool calls converge on shared databases or caches, while automatic multi-step retries can amplify failures. The excerpt identifies three patterns: synchronized writes when sub-agents finish together, semantically similar cache misses after correlated expiry, and workflow-wide retry storms. A Cockroach Labs benchmark reported a sharp saturation inflection between 700 and 1,000 concurrent agents; at 5,000 and 10,000 agents, PostgreSQL throughput fell below CockroachDB's. The author notes the benchmark is vendor-run and advises testing representative workloads. Suggested controls include staggering agent dispatch and cache TTLs with jitter, sizing connection pools for fan-out, triggering circuit breakers on behavioral signals such as latency growth and write volume, avoiding sequential-key hotspots, and using sufficiently granular durable checkpoints. The excerpt cautions that results at hundreds of agents may not predict behavior at thousands.

### Source excerpt

The thundering herd of the past was externally triggered.

## Agentic AI Architecture: How CockroachDB Supports Memory, Context, and Control

DevFeed: [Agentic AI Architecture: How CockroachDB Supports Memory, Context, and Control](<https://devfeed.tech/articles/agentic-ai-architecture-how-cockroachdb-supports-memory-context-and-control-23734.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/agentic-ai-architecture-memory-control>)

Author: Alejandro Infanzon

Published: 2026-06-11T00:00:00Z

Content type: article

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [agentic workflows](<https://devfeed.tech/topics/agentic-workflows.md>), [Database](<https://devfeed.tech/topics/database.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [LLM security](<https://devfeed.tech/topics/llm-security.md>), [context window](<https://devfeed.tech/topics/context-window.md>), [Python](<https://devfeed.tech/topics/python.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agentic-ai-architecture](<https://devfeed.tech/tags/agentic-ai-architecture.md>), [agentic-systems](<https://devfeed.tech/tags/agentic-systems.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agent-memory](<https://devfeed.tech/tags/ai-agent-memory.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-architecture](<https://devfeed.tech/tags/ai-architecture.md>), [ai-observability](<https://devfeed.tech/tags/ai-observability.md>), [ai-workloads](<https://devfeed.tech/tags/ai-workloads.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [autonomous-ai-agents](<https://devfeed.tech/tags/autonomous-ai-agents.md>), [availability](<https://devfeed.tech/tags/availability.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [context](<https://devfeed.tech/tags/context.md>), [control](<https://devfeed.tech/tags/control.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [memory](<https://devfeed.tech/tags/memory.md>)

### AI overview

The article presents a CockroachDB-based multi-agent prototype that stores durable agent state, prior analyses, telemetry and audit metadata alongside business data. Its LangGraph swarm uses a CockroachDB MCP server to inspect live schema and cluster health before SQL work, and native vector search to find similar saved reports; users can reuse a match or refresh the analysis. A demo describes a similar query returning an 85.78% match and bypassing the planner, SQL and web research steps. Python decorators record agent timing and usage metrics, while IAM scopes database execution to the user's session. The article reports that telemetry exposed an 11-minute Critic loop, which termination guards and instruction stripping reduced to about 40 seconds on measured runs. It frames these mechanisms as ways to reduce repeated work and improve cost visibility and auditability, while describing CockroachDB's distributed architecture as supporting resilience under high concurrency.

### Source excerpt

What happens when you connect a fleet of autonomous AI agents to your enterprise data stack? You quickly discover...

## The Bill Arrives: How to Manage Agentic AI Costs at Scale

DevFeed: [The Bill Arrives: How to Manage Agentic AI Costs at Scale](<https://devfeed.tech/articles/the-bill-arrives-how-to-manage-agentic-ai-costs-at-scale-23736.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/agentic-ai-costs-at-scale>)

Author: Quentin Packard

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

Content type: article

Language: en

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

Topics: [Finance](<https://devfeed.tech/topics/finance.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [Loop Engineering](<https://devfeed.tech/topics/loop-engineering.md>), [Claude](<https://devfeed.tech/topics/claude.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [database-optimization](<https://devfeed.tech/topics/database-optimization.md>), [consistency](<https://devfeed.tech/topics/consistency.md>), [ai-governance](<https://devfeed.tech/topics/ai-governance.md>), [managing large-scale text data in AI](<https://devfeed.tech/topics/managing-large-scale-text-data-in-ai.md>), [Local AI](<https://devfeed.tech/topics/local-ai.md>), [model architecture](<https://devfeed.tech/topics/model-architecture.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.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-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-costs](<https://devfeed.tech/tags/ai-costs.md>), [bill](<https://devfeed.tech/tags/bill.md>), [budget](<https://devfeed.tech/tags/budget.md>), [building](<https://devfeed.tech/tags/building.md>), [business](<https://devfeed.tech/tags/business.md>), [context](<https://devfeed.tech/tags/context.md>), [costs](<https://devfeed.tech/tags/costs.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [manage](<https://devfeed.tech/tags/manage.md>), [production](<https://devfeed.tech/tags/production.md>), [scale](<https://devfeed.tech/tags/scale.md>), [token](<https://devfeed.tech/tags/token.md>), [uber](<https://devfeed.tech/tags/uber.md>)

### AI overview

Agentic AI costs depend on the full cost of completing a task, not just model token prices: multi-step workflows resend context, retrieve data, call tools, retry failures, and require orchestration, evaluation, and governance. The article says redundant context and context growth can raise inference costs while degrading output quality; it recommends compaction, just-in-time retrieval, focused tool sets, and isolated sub-agents. RAG adds embedding, vector database, re-indexing, and data-preparation costs, but can be cheaper and faster than sending very large contexts at high query volumes; long context may cost less for low-volume internal use. It highlights prompt caching as a high-return cost lever, provided cached prefixes avoid changing timestamps, IDs, and other dynamic content. For production control, it recommends monitoring usage alongside outcomes such as tasks completed, measuring value per token, and limiting runaway retries with human escalation. Database latency and inconsistent state can trigger retries that resend context and increase spend. The figures and case studies are claims made in the supplied article.

### Source excerpt

What do the Uber budget blowout, a 24x token multiplier, and context teach us about building a real business case for AI Agents in production?

## RoachFest London 2026 Preview: AI Agents and an Astronaut Walk Into a Database Conference

DevFeed: [RoachFest London 2026 Preview: AI Agents and an Astronaut Walk Into a Database Conference](<https://devfeed.tech/articles/roachfest-london-2026-preview-ai-agents-and-an-astronaut-walk-into-a-database-conference-23808.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/roachfest-london-2026-preview>)

Author: Rob Reid

Published: 2026-06-09T00:00:00Z

Content type: news

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [Database](<https://devfeed.tech/topics/database.md>), [Transactions](<https://devfeed.tech/topics/transactions.md>), [acid](<https://devfeed.tech/topics/acid.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [astronaut](<https://devfeed.tech/tags/astronaut.md>), [conference](<https://devfeed.tech/tags/conference.md>), [database](<https://devfeed.tech/tags/database.md>), [london](<https://devfeed.tech/tags/london.md>), [preview](<https://devfeed.tech/tags/preview.md>)

### AI overview

RoachFest London 2026 is scheduled for June 25 at Convene, with a program focused on resilience, distributed databases, and adapting to AI. The announced sessions include CockroachDB CEO Spencer Kimball on database architecture as agentic AI changes scale; Form3's cloud-agnostic active/active/active payment platform across AWS, GCP, and Azure, whose disaster recovery is tested by taking a provider offline for 24 hours in production; and a panel on AI-agent state tradeoffs including semantic retrieval and SQL-native memory. Workshops cover distributed SQL, Raft, survivability and multi-region strategies, plus transactional vector search and RAG on CockroachDB. The event also features product roadmap topics such as BYOC and AI-powered migration, partner demos, and astronaut Tim Peake on preparation and resilience.

### Source excerpt

Cockroach Labs has been hosting our annual database conference since 2022, and I'm honored to be MCing RoachFest London for the third year running.

## What Breaks When Agentic AI Reaches Production?

DevFeed: [What Breaks When Agentic AI Reaches Production?](<https://devfeed.tech/articles/what-breaks-when-agentic-ai-reaches-production-23738.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/agentic-ai-production-infrastructure>)

Author: Quentin Packard

Published: 2026-06-04T00:00:00Z

Content type: article

Language: en

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

Topics: [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Production Engineering](<https://devfeed.tech/topics/production-engineering.md>), [Monitoring](<https://devfeed.tech/topics/monitoring.md>), [AI Context Management](<https://devfeed.tech/topics/ai-context-management.md>), [agent orchestration](<https://devfeed.tech/topics/agent-orchestration.md>), [incident](<https://devfeed.tech/topics/incident.md>), [site-reliability-engineering](<https://devfeed.tech/topics/site-reliability-engineering.md>), [context window](<https://devfeed.tech/topics/context-window.md>)

Tags: [2](<https://devfeed.tech/tags/2.md>), [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [account](<https://devfeed.tech/tags/account.md>), [actions](<https://devfeed.tech/tags/actions.md>), [adaptive](<https://devfeed.tech/tags/adaptive.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [ai](<https://devfeed.tech/tags/ai.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [enterprise-ai](<https://devfeed.tech/tags/enterprise-ai.md>), [fewer](<https://devfeed.tech/tags/fewer.md>), [incident](<https://devfeed.tech/tags/incident.md>), [one](<https://devfeed.tech/tags/one.md>), [production](<https://devfeed.tech/tags/production.md>), [program](<https://devfeed.tech/tags/program.md>), [question](<https://devfeed.tech/tags/question.md>), [teams](<https://devfeed.tech/tags/teams.md>)

### AI overview

The article argues that production agent failures usually stem from distributed-systems and infrastructure gaps, not the model. It identifies six challenges: durable memory across node failures and concurrent updates; machine-speed fan-out that can cause cache stampedes, retry storms, and shared-state hotspots; agent identity with scoped, rotatable credentials and auditable delegation; blast-radius controls such as circuit breakers; durable, structured execution records for investigation and compliance; and unit costs that can rise as sessions and histories grow. It recommends patterns including connection pooling, adaptive rate limits, jittered retries, data-layer permission boundaries, behavioral thresholds, and modeling realistic per-session costs before launch. It presents these as engineering requirements for external deployment, while noting that they require upfront investment.

### Source excerpt

Most enterprise AI teams have built an agent that was impressive; far fewer have shipped one without a production incident that made someone question the whole program.

## How to Choose a Database for an AI-Powered Product

DevFeed: [How to Choose a Database for an AI-Powered Product](<https://devfeed.tech/articles/how-to-choose-a-database-for-an-ai-powered-product-23775.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/database-for-ai-applications>)

Author: David Weiss

Published: 2026-06-01T00:00:00Z

Content type: tutorial

Language: en

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

Topics: [Database](<https://devfeed.tech/topics/database.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [AI Platform](<https://devfeed.tech/topics/ai-platform.md>), [Data Infrastructure](<https://devfeed.tech/topics/data-infrastructure.md>), [autoscaling](<https://devfeed.tech/topics/autoscaling.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-infrastructure](<https://devfeed.tech/tags/ai-infrastructure.md>), [capacity](<https://devfeed.tech/tags/capacity.md>), [concurrency](<https://devfeed.tech/tags/concurrency.md>), [data](<https://devfeed.tech/tags/data.md>), [database](<https://devfeed.tech/tags/database.md>), [how-to](<https://devfeed.tech/tags/how-to.md>), [product](<https://devfeed.tech/tags/product.md>), [scale](<https://devfeed.tech/tags/scale.md>)

### AI overview

This article presents seven criteria for choosing a database for an AI-powered product in production. It emphasizes evaluating elastic scalability, concurrency, correctness, workload demands, global distribution, and access by autonomous agents rather than selecting a database solely for model support or vector search.

### Source excerpt

Organizations across fintech, healthcare, retail, gaming, SaaS, and scores more verticals are embedding AI into business-critical offerings.

## AI Agents Need Context to Reason, Not Just Data

DevFeed: [AI Agents Need Context to Reason, Not Just Data](<https://devfeed.tech/articles/ai-agents-need-context-to-reason-not-just-data-23742.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/ai-agent-context-management>)

Author: Quentin Packard

Published: 2026-05-28T00:00:00Z

Content type: article

Language: en

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

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [context](<https://devfeed.tech/topics/context.md>), [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [Agentic AI](<https://devfeed.tech/topics/what-is-agentic-ai.md>), [Data Infrastructure](<https://devfeed.tech/topics/data-infrastructure.md>)

Tags: [agentic-ai](<https://devfeed.tech/tags/agentic-ai.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-architecture](<https://devfeed.tech/tags/ai-architecture.md>), [context](<https://devfeed.tech/tags/context.md>), [data-infrastructure](<https://devfeed.tech/tags/data-infrastructure.md>), [database](<https://devfeed.tech/tags/database.md>)

### AI overview

The article argues that production failures in AI agents often stem from inadequate context management rather than the model itself. Reliable agent behavior requires current data, memory, permissions, observability, and awareness of system constraints, making context management a data infrastructure problem beyond basic retrieval or prompt engineering.

### Source excerpt

When your AI agent makes a bad decision in production, what do you blame?

## Scalable Leader Leases for Distributed SQL Databases: CockroachDB at SIGMOD 2026

DevFeed: [Scalable Leader Leases for Distributed SQL Databases: CockroachDB at SIGMOD 2026](<https://devfeed.tech/articles/scalable-leader-leases-for-distributed-sql-databases-cockroachdb-at-sigmod-2026-23779.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/distributed-database-leader-leases>)

Author: Rebecca Taft

Published: 2026-05-26T00:00:00Z

Content type: news

Language: en

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

Topics: [CockroachDB](<https://devfeed.tech/topics/cockroachdb.md>), [distributed algorithm](<https://devfeed.tech/topics/distributed-algorithm.md>), [Raft](<https://devfeed.tech/topics/raft.md>), [scalable-architecture](<https://devfeed.tech/topics/scalable-architecture.md>), [consensus algorithm](<https://devfeed.tech/topics/consensus-algorithm.md>), [Database](<https://devfeed.tech/topics/database.md>), [sharding](<https://devfeed.tech/topics/sharding.md>), [Cockroach Labs](<https://devfeed.tech/topics/cockroach-labs.md>), [health check](<https://devfeed.tech/topics/health-check.md>)

Tags: [2022](<https://devfeed.tech/tags/2022.md>), [2025](<https://devfeed.tech/tags/2025.md>), [2026](<https://devfeed.tech/tags/2026.md>), [clusters](<https://devfeed.tech/tags/clusters.md>), [cockroach-labs](<https://devfeed.tech/tags/cockroach-labs.md>), [cockroachdb](<https://devfeed.tech/tags/cockroachdb.md>), [conference](<https://devfeed.tech/tags/conference.md>), [consensus](<https://devfeed.tech/tags/consensus.md>), [cost](<https://devfeed.tech/tags/cost.md>), [cpu](<https://devfeed.tech/tags/cpu.md>), [crashes](<https://devfeed.tech/tags/crashes.md>), [data-management](<https://devfeed.tech/tags/data-management.md>), [databases](<https://devfeed.tech/tags/databases.md>), [decentralized](<https://devfeed.tech/tags/decentralized.md>), [distributed](<https://devfeed.tech/tags/distributed.md>), [distributed-databases](<https://devfeed.tech/tags/distributed-databases.md>), [distributed-sql](<https://devfeed.tech/tags/distributed-sql.md>), [leader](<https://devfeed.tech/tags/leader.md>), [paper](<https://devfeed.tech/tags/paper.md>), [sql](<https://devfeed.tech/tags/sql.md>)

### AI overview

CockroachDB's proposed Leader Leases combine decentralized node health checks with Raft leader fortification so each Range's leaseholder can also be its Raft leader. The approach targets the CPU cost of expiration-based leases, which renew through replicated Raft writes per Range, while avoiding centralized leases' dependence on one Range for cluster health status. Its three components are the Liveness Fabric, Leader Fortification, and the unified Leader Leases abstraction. The article reports up to 85% lower CPU usage for lease maintenance at scale, stable throughput as Range counts grew, and recovery within seconds under tested node crashes, network partitions, and disk stalls. It says the Liveness Fabric's safety properties were formally verified in TLA+, with the model publicly available. These results are reported by the paper's authors; the excerpt does not provide experiment details.

### Source excerpt

We are pleased to announce that Cockroach Labs has a new paper appearing at SIGMOD 2026: Scalable Leader Leases For Multi Consensus Groups in CockroachDB.

[Next page](<https://devfeed.tech/sources/cockroach-labs.md?cursor=WyIyMDI2LTA1LTI2VDAwOjAwOjAwKzAwOjAwIiwgIjA1Y2QxYTVkLTdlM2UtNDdkMi1iNGQ2LTY3MjY1YTI3OGQ1YSJd>)