How We Cut API Latency by More Than Half with PostgreSQL and RedisLatest
Engineering

How We Cut API Latency by More Than Half with PostgreSQL and Redis

Our /call API took 700-800 ms at p95. Removing redundant PostgreSQL queries, caching slow-changing data in Redis, running independent reads concurrently and dropping needless transactions brought it to 300-400 ms, with no new infrastructure.

Read
Read How We Cut API Latency by More Than Half with PostgreSQL and Redis
How We Built a Read-Only AI Agent for Root Cause Analysis in Production
Engineering

How We Built a Read-Only AI Agent for Root Cause Analysis in Production

An evidence-first architecture for AI root cause analysis in production: how RootCauseClaude, our read-only debugging agent in Slack, investigates incidents with code, logs, traces and Git-based memory, usually within two to ten minutes.

Read
Read How We Built a Read-Only AI Agent for Root Cause Analysis in Production
Testing Jev on real phone calls
Engineering

Testing Jev on real phone calls

TypeSafe's Jev is pitched at exactly the decisions that sit around a phone call. We put it on three real jobs against 17 models: it is 1.7 to 3.3 times faster everywhere, level with the best on judging and extraction, and not ready for routing.

Read
Read Testing Jev on real phone calls