Writing
#llm
2 articles tagged llm.
Building Forge: Postgres-Orchestrated LLM Batch and Realtime Jobs Without a Queue
Forge runs LLM batch and realtime jobs across OpenAI and Anthropic with two Node daemons that share nothing but a Postgres database. No queue, no IPC, no shared worker process — the database does the coordinating. Here is how that architecture works and what I learned building it.
aillminfrastructurepostgresarchitecture
Evaluation-Driven Development: Shipping an AI Booking Agent You Can Trust
I was fixing Matcha's booking agent one conversation at a time, with no metric to tell me whether a change helped or quietly regressed. The fix: evaluate the whole agentic flow with openevals — golden fixtures, multi-turn simulated users, an adversarial safety floor — and promote a change only when the numbers say it's better.
aiai-agentsevaluationtestingllm