Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
How short-term (thread-scoped) and long-term (cross-thread) memory actually work in LangGraph, with code, schemas, and the eviction policies that keep cost predictable.
Use LangGraph's checkpointer to make agents resumable across crashes and human-in-the-loop pauses, then replay any checkpoint into your eval pipeline.
How LangGraph's StateGraph, channels, and reducers actually work — with a working multi-step agent, eval hooks at every node, and the patterns that survive production.
Langgraph multi-agent supervisor handoffs docs: the supervisor pattern in LangGraph for coordinating specialist agents, with full code, an eval pipeline that scores routing accuracy, and the failure modes to watch for.
Pairwise (A vs B) LLM-as-judge evaluation produces sharper, more reliable signal than absolute scoring for non-deterministic agent outputs. Here is why and how.
Handoffs done right — when one agent should hand control to another, how to preserve context, and how to evaluate the handoff decision itself.
Offline evals catch regressions before deploy on a fixed dataset. Online evals catch real-world drift on live traffic. You need both — here is how we run them.
Honest principal-engineer comparison of the OpenAI Agents SDK and the legacy Assistants API, with a migration checklist and eval-parity strategy so you don't ship regressions.
Build a working computer-use agent with the OpenAI Computer Use tool — clicks, types, scrolls a real browser — then evaluate task success on a benchmark suite.
