All posts
Long-form notes on software, distributed systems, and the craft of building. Shipping one a week.
The Deterministic Backbone: Why Production AI Systems Are Moving Away From Fully Autonomous Agents
Fully autonomous agents are hard to bound, hard to test, and expensive to operate. A deterministic backbone with narrow agent steps gives you the control flow back while keeping the intelligence where it matters. Here is how to design, test, and migrate toward it.
Memory Evaluation: Measuring How AI Memory Decays Over a Project's Lifetime
Most AI memory benchmarks grade on recall and stop there. That hides the real failure mode: stale facts quietly poisoning the context window. Here is a lifecycle-based evaluation framework that tests recall, revision, and controlled forgetting across the change points every long-lived project goes through.
The Transactional Outbox Is Not a Queue
The transactional outbox is a ledger, not a queue. Treating it like one is what breaks Postgres under load. This post walks through the specific failure modes — autovacuum stalls, xmin horizon drift, replication slot lag, poison pills — and the operational rules that actually keep it working in production.
Virtual Threads After JEP 491: The Bottleneck Moved
JEP 491 removed the `synchronized` pinning problem that kept virtual threads out of production. The interesting question now isn't whether to enable them — it's which bottleneck shows up next. A field guide for Spring Boot / Kotlin services running on JDK 24+.