Topic
All digests tagged LLM Agents
Suggested route
Move from earlier context through the available material to the newest signals. This route is a reading aid, not a difficulty ranking.
Harness engineering is the infrastructure around LLMs, encompassing tools, context, guardrails, and telemetry. While LLMs introduce non-determinism, core software principles like CI/CD, security, and observability remain essential for building robust agents.
Read article
Learn how to move from subjective 'vibe checks' to systematic AI evaluations (evals). This guide covers defining concrete pass/fail criteria, analyzing agent traces, and building robust eval templates using LLM judges and Decision Models.
Read article
Browser agents are essential for accessing financial data behind logins or anti-bot walls. Browserbase demonstrates how forking Chromium and using Stagehand allows for scalable, reliable automation across thousands of different vendor portals.
Read article
CoreWeave Agent Lens delivers intelligent agent observability at enterprise scale, providing a single view of every trace to turn production issues into proven fixes.
Read article
Bigger context windows won't save your AI agents. Learn why models lose information in the middle and master the four strategies of Context Engineering: Externalize, Select, Compress, and Isolate.
Read article
Comparing agentic search vs. vector search for coding agents, this talk reveals that while both methods achieved equal accuracy, vector search cost four times more due to lack of connective context.
Read article
LangSmith is LangChain's platform for the Agent Development Lifecycle (ADLC). It provides critical observability, allowing engineers to trace complex LLM agent interactions, debug tool calls, and monitor performance in production.
Read article
The talk details 'harness engineering,' arguing that advanced AI skills must move beyond simple prompting. Techniques include using blind sub-agents, anti-attractors, and hooks to build robust, deterministic, and multi-model compatible systems.
Read article
Learn how middleware extends Managed Deep Agents, enabling custom behaviors like PII redaction and tool call logging. See how to build custom middleware using `wrap_tool_call`.
Read article
Formal verification is necessary for GPU kernels because testing fails due to non-determinism and massive input spaces. The solution involves using SMT solvers to prove algorithmic correctness at the PTX level.
Read article
Learn how to analyze thousands of production traces and build structured datasets of agent failures in minutes. This workflow uses the LangSmith CLI and a coding agent to classify perceived errors into reusable failure types.
Read article
Context engineering is critical for LLM agents. The talk reveals that compaction techniques can be detrimental because they invalidate prompt caches, forcing full-price token usage. Full history retention often wins on recall.
Read article
OpenWiki is an open-source CLI that generates documentation specifically optimized for coding agents. It uses OKF and GitHub Actions to automatically maintain docs synced with your repo's Git history, improving agent efficiency.
Read article
Refactoring a complex, multi-repo codebase into a monorepo was deemed worthwhile despite AI progress. Benchmarks show modern LLMs drastically reducing refactor time, but manual validation remains crucial.
Read article
Mastering the Agent Development Lifecycle requires moving beyond demos to systematic iteration across Build, Test, Deploy, Monitor, and Govern. Learn how advanced tooling uses tracing and online evals to ensure agents behave reliably at scale.
Read article
Learn how to trace every Cursor agent turn into LangSmith by installing the plugin, setting three environment variables, and running an end-to-end task.
Read article