Essential Read10 min read
The AI Engineering Roadmap: From Script to Scale
Stop building "demos" and start building "products". A comprehensive guide to the stack, the skills, and the mindset shift required.
Read the full guide →
I write about the messy reality of building production AI systems. No hype, just lessons learned from shipping code.
No spam. Just shipping stories.
Stop building "demos" and start building "products". A comprehensive guide to the stack, the skills, and the mindset shift required.
Tutorials, opinion pieces, and release notes.
Most RAG prototypes cost $2-5 per query. At 10,000 queries per day, that is $1.5M per year. I cut retrieval costs by 99% in production. Here is the playbook.
We don't deploy code without tests. Why are we deploying AI without evals? A practical guide to building evaluation harnesses that make AI systems reliable enough to trust.
Vendor lock-in in AI is existential. Here is how I architected a vendor abstraction layer that cut a client's AI spend from $78K/mo to $18K/mo — and made their infrastructure antifragile in the process.
Why the biggest risk in AI isn't the model — it's everything around it. How a production engineering mindset turns fragile demos into hardened production systems.
A deep dive into the latency budget of voice AI pipelines, why WebRTC changes everything, and how I built a production voice agent on celestino.ai that responds in under a second.
A manifesto on why AI-native engineering is the future of software development.
Ask questions about these topics, my engineering approach, and how I build AI-native products.