Insights
Thoughts, observations, and lessons from building practical AI systems.
Last updated April 2026
Building Agentic Systems That Don't Frustrate Users
Most autonomous AI agents try to do too much. Here's why keeping humans in the loop usually works better for real-world problems.
Why More Companies Are Choosing Private AI Deployments
Data privacy, compliance concerns, and rising API costs are driving more organizations to run AI models on their own infrastructure.
RAG in the Real World: Lessons From Three Production Projects
Retrieval-augmented generation sounds simple in theory, but there are a lot of subtle details that make it work well in practice.
Designing Human-in-the-Loop AI Systems
What we've learned about building AI systems that work with people instead of trying to replace them.
The Case for Simpler AI Infrastructure
Do you really need Kubernetes and a dozen different microservices to run AI in production? Not always.
Thinking About AI Agent Evaluation
Measuring whether an AI agent is actually doing a good job is harder than it looks. Some approaches we've found useful.