API Versioning in Go: URL Strategies, Breaking Changes, and Deprecation That Doesn’t Break People
One Tuesday your API’s most popular endpoint works fine. Two Tuesdays later, a mobile app that has been calling it
From Our Blog
Distributed Tracing in Go with OpenTelemetry: Spans, Context, and Sampling That Works
Your microservice takes eleven seconds to answer a request that used to take two hundred milliseconds. The on-call engineer opens
MHA, MQA, GQA, MLA: The KV-Cache Trade-offs Behind Every Modern Transformer
Ask someone to sketch a transformer on a whiteboard and they’ll draw the 2017 version: softmax attention over every pair
Continue readingMHA, MQA, GQA, MLA: The KV-Cache Trade-offs Behind Every Modern Transformer
Distributed Rate Limiting in Go: GCRA, Redis, and the Pitfalls Nobody Warns You About
Every production rate limiter has two layers. The first layer is the algorithm: token bucket, leaky bucket, fixed window, sliding
Reading EXPLAIN Like a Planner: Debugging Slow Queries in PostgreSQL
# POST 2: Reading EXPLAIN A query that ran in 20 milliseconds yesterday takes 40 seconds today. Nothing in the
Continue readingReading EXPLAIN Like a Planner: Debugging Slow Queries in PostgreSQL
The Transactional Outbox Pattern in Go: Reliable Events Without Dual Writes
# POST 1: Transactional Outbox (Slot 10 backend) You update a row in your database, then publish an event to
Continue readingThe Transactional Outbox Pattern in Go: Reliable Events Without Dual Writes
npm vs Yarn vs pnpm vs Bun: Fifteen Years of Package-Manager Wars, as a Timeline
Four package managers have shaped a decade of JavaScript tooling. npm is the incumbent that ships with Node, Yarn was
Continue readingnpm vs Yarn vs pnpm vs Bun: Fifteen Years of Package-Manager Wars, as a Timeline
Every 2026 LLM Release, Paired With the Hype That Announced It — Scored for Accuracy
What this is: a chronology of the major LLM releases of 2026, paired with the loudest things the internet said
Continue readingEvery 2026 LLM Release, Paired With the Hype That Announced It — Scored for Accuracy
Stop Vibing Your RAG Pipeline: Recall@k, MRR, and NDCG in an Afternoon
Most teams that ship a retrieval-augmented generation feature never measure it. The demo looks good, a few hand-picked questions return
Continue readingStop Vibing Your RAG Pipeline: Recall@k, MRR, and NDCG in an Afternoon
Structured Logging in Go With slog: Handlers, Sampling, and Dynamic Levels
A request fails at 2 AM. The on-call engineer greps through logs looking for a trace ID, finds seventeen lines
Continue readingStructured Logging in Go With slog: Handlers, Sampling, and Dynamic Levels