LangSmith Engine now detects agent issues over 2x better, proposes stronger fixes, supports Slack and Linear workflows, and is available for self-hosted deployments.

Incentives favor offense when it comes to agentic cybersecurity; it’s the same dynamic that will limit incumbents and fuel startups in the long run.

Vercel Connect is now generally available. Agents request short-lived, scoped tokens at runtime instead of storing provider secrets that never expire.

See how Toyota North America uses Deep Agents and LangSmith to run 50+ production agents, cut delivery from 6 months to 4 days, and track AI ROI.

New on-silicon performance data measured by NVIDIA using real-world agentic coding trajectories shows Vera Rubin NVL72 systems deliver 30x higher throughput per megawatt and 35x lower token costs than NVIDIA GB300 NVL72.

A JPEG XL decoder is heading for Firefox 157! Here's how we're shipping it safely…

An append-only status model gives you full history, but concurrent transitions can produce contradictory records. Here’s how to prevent that.

Agent evaluation tooling is shipping fast, and the message is clear: production quality depends on your evaluation pipeline, not just your prompt. This week ...

AI is accelerating software delivery, but regulated industries face a harder challenge: moving faster without increasing risk. Here’s how teams can balance speed with privacy, security, compliance, and accountability.

AI makes code faster and cheaper to produce. Premium consultancies need to offer more: experienced judgment, the right capabilities, and people who help teams solve the right problems and make better decisions.

Cursors. Temp tables. Multi-statement transactions. The procedural code your data warehouse runs on now lives on the lakehouse – translated, not rewritten.

Preview Builds let teams test pull request branches in temporary, production-like LangSmith deployments before merging agent changes.

Today's maybe more of a "representative data sheet entry" than anything else. Every developer has the experience of reading the documentation. If you've been at this for some time, you've probably read bad documentation. Documentation that is incomplete, inaccurate, or otherwise flawed. Or, my personal favorite, the brief time where Oracle tried to put all of its documentation into an Adobe Flex site (aka, a Flash application, not a real web app). That one had fun bonus features, like "breaking copy and paste" and "preventing you from deep linking to a piece of the documentation".

Go 1.27 adds generic methods, encoding/json/v2 package, uuid package, faster memory allocation, goroutine leak profiles, and more.

T.J. Maher, an automation developer since 2015, blogs about his transition from a manual tester to a software engineer in test.

LangSmith Tuned Evaluators attach quality feedback to production traces, starting with Perceived Error, to help teams find and fix agent mistakes.

Lessons from the Grounded Reasoning Cup, where leading academic teams evaluated AI agents live on a newly released enterprise grounded-reasoning benchmark.