PostgreSQL is still treated like 'just the database,' even though it can absorb surprising amounts of search, scheduling, vector, auth-adjacent, and API work. OpenAI's own PostgreSQL scaling story is a good reminder: this tool has far more headroom than most teams assume.
Notes
Engineering notebook
Notes on runtime behavior, failure modes, deployment tradeoffs, and the implementation details that still matter after the demo ends.
01 / Notes
Notebook entries
Notes and related entries, ordered by date.
Engineering
Building CoreLink AI: An Evidence-Grounded Reasoning Engine That Knows When to Search, Compute, and Stop
Most agent systems fail in predictable ways: they trust model recall too much, use tools without policy, and keep reasoning long after the evidence ran out. CoreLink AI was built to make those failure modes explicit, bounded, and auditable.
Production failures in computer vision are rarely simple 'wrong predictions.' They are complex conceptual drifts—blur, lighting, camera shifts. Here’s how I built a 'flight recorder' to catch them before they become incidents.
December isn’t “dead.” It’s noisy, uneven, and full of timing signals. Here’s what I observed by tracking company hiring momentum daily—and how to prepare for January after the holidays.
Job Search
The Hiring Momentum Dashboard I Wish Existed
I stopped tracking individual job listings and started tracking hiring behavior—momentum, freezes, job lifespan, and timing. Here’s what changed and why it’s actually useful.