2026-09-03 · notes
Harness engineering
A harness is executable knowledge: capture, replay, assert, retain.
Writing
Notes from building real systems
Working notes on memory, reproducibility, integrations, infrastructure, and what production teaches us about AI engineering.
2026-09-02 · notes
Reproducible debugging for AI-assisted engineering
Agents are only as useful as the feedback loop you give them. Make the failure stand still.
2026-09-01 · notes
Why agent memory is an infrastructure problem
Chat logs and vector dumps are not memory. Durability, schema, provenance, and evaluation are.
2026-08-20 · notes
AI agents do not remove systems engineering
Tool calling is a new I/O surface. Queues, identity, permissions, and failure handling remain the job.
2026-08-12 · notes
Context as infrastructure
Retrieval, boundaries, and exclusions are as load-bearing as a schema. Treat them that way.
2026-07-30 · notes
Building reliable integrations
Webhooks lie, clocks disagree, and identity is a negotiation. Reliability is idempotency plus evidence.
2026-07-15 · notes
Engineering lessons from production AI systems
A short list of lessons that survive contact with production. No invented metrics.
2026-06-20 · notes
Multi-agent software development
More agents are not more leverage unless the interfaces are boring and the tests are real.
2026-06-01 · notes
Infrastructure and security notes
Reliability and security share a root: defined boundaries, least privilege, and evidence you can audit.