Kyle Stratis is the founder of Stratis Data Labs, VP of engineering at Telepath, and the author of AI Agents with MCP (O’Reilly Media). Over more than a decade in AI engineering, computer vision, and machine learning, he has built AI platforms, data pipelines, and cloud analytics systems for both startups and large enterprises. He has been implementing Model Context Protocol clients and servers since early 2025, and runs MCP workshops at ODSC conferences and custom ones for clients.
His specialty is the gap between a promising experiment and a shipped product. At Vizit Labs he led R&D engineering, tracking down the bottlenecks in the model development process to cut training time from a full day to 15 minutes, then building the automated training pipeline, feature store, and model definition composer that let non-engineering teams deliver custom models on their own. Since going independent he has built and evaluated agents for clients including Microsoft, LogicMonitor, and Koch Industries, contributed major features to Tribe AI’s enterprise agent framework, and taught MCP in Python to engineering teams at Nelnet.
He writes about Python, LLMs, and personal knowledge management at kylestratis.com, and covers AI news at The Signal Path.
Speaking
- From Intelligent to Agentic Applications: Using Model Context Protocol to Support Agentic Behaviors in your Applications — ODSC Agent AI Summit (2026), ODSC West (2025)
- Towards a Truly End-to-End Machine Learning Pipeline — Emotion AI Summit (2019)
Interested in having Kyle speak at your conference or lead a workshop? Get in touch.
