Global scale
I run AI where trust, adoption, and economics all matter.
Five concurrent enterprise programs, a $30M portfolio, 120+ contributors, and CEO-active platforms across a global public company.
D. Brian Letort, Ph.D.
AI strategy for people who have to operate it.
I write about the systems layer of enterprise AI: the models, data foundations, agentic workflows, economics, governance, and operating patterns that turn experiments into durable capability.
- Global enterprise AI operator
- $30M AI and data portfolio
- Hands-on builder and coder
- AI industry intelligence
- PhD-level teaching
- 1,000+ member AI community
- 5 U.S. Patents
- Context Compilation trilogy
- 16 Pluralsight courses
- 2 books
Follow the operating layer
Board-ready AI strategy, operator-grade technical depth, and weekly signal from the AI market.
Start reading
What to read next.
Industry intelligence
AI market signal, organized weekly.
Cross-stack flywheel
AI Stack Weekly
For officers tracking AI market movement.
Latest / Issue 14
The mid-tier compressed frontier economics while the rack-and-power layer turned into a two-vendor race
Model layer
The Model Pulse
For architects tracking model capability shifts.
Latest / Issue 14
Opus-class intelligence just moved into the mid-tier price band, while Google reset the fleet-economics floor
Operating layer
Agent Techniques Weekly
For builders operationalizing agentic work.
Latest / Issue 14
Fallback routing moved from a homegrown harness trick into the model API — and that makes provenance, not uptime, the hard problem
Procurement surface
The Application Layer
For buyers watching AI reshape software.
Latest / Issue 13
The model API became an agent control plane: fallbacks, mutable tools, and fleet economics moved into the runtime
Research
Context Compilation Theory.
A foundational paper and three-paper trilogy on the systems layer between retrieval and reasoning — the governed, measurable, optimizable layer where reliability lives, where provenance is enforced, and where token economics becomes a design surface rather than an accounting line.
Why trust the signal
Strategic altitude. Technical depth. Global scale.
Technical depth
I can move from boardroom framing to code-level architecture.
Data, data science, machine learning, software engineering, agents, RAG, model routing, evals, and the operating layer between them.
Public body of work
I write, teach, publish, and build in the open.
A Context Compilation research program, PhD teaching, 16 Pluralsight courses, two books, and a weekly AI industry corpus.
Where to go next
Choose your path.
Speaking and executive briefings
I accept a small number of speaking, board, press, and executive briefing requests each year.