Transmission Log
Lab Notes
Technical dispatches from the R&D lab. Engineering deep dives, AI strategy, and lessons from production.
Engineering
Zero Parse Failures at Scale: Structured Output with Local LLMs
We validated Ollama's grammar-constrained structured output at scale — 557 consecutive calls, zero parse failures. On-premises AI pipelines now get the same reliability guarantees as cloud APIs.
Read
AI Strategy
Why Most AI Projects Fail (And How We Prevent It)
After 15 years of building production systems, here's what actually makes AI projects succeed.
Read
Case Study
Automating 100 Daily Physician Referrals with AI
A sleep medicine practice was losing referrals to a manual fax-based intake pipeline. We built a document processing system that handles them at 98% accuracy.
Read