Writing
I write about how AI changes what we build, how companies operate, and where lasting advantage comes from. My perspective combines product management, engineering, and strategy, grounded in building systems and leading teams.
18 pieces
The Evolution of Engineering Interviews: A Few Thoughts
AI should be part of engineering interviews, alongside fundamentals, system design, evaluation, debugging, and production ownership.
The AI-Native Data Warehouse: Keep Truth Atomic, Make Aggregation Disposable
AI-native analytics needs atomic facts as the source of truth, governed semantic context, and pre-aggregated models as optional accelerators.
Why Evals Matter in AI Analytics
A video walkthrough of evaluation in AI analytics, from checking a churn metric to explaining its causes and deciding what to do.
The Semantic Layer and Data Modeling Dilemma in AI Analytics
AI is driving self-serve data adoption, but accuracy depends on balancing core data modeling, curated context, and continuous evaluation.
AI Does the Work. People Decide What Matters
AI is removing traditional junior work while allowing junior people to contribute earlier. The real challenge is rebuilding judgment, apprenticeship, and accountability around that shift.
Building an AI-Native Organization: Process, Technology, People
Adoption is easy to count. Compounding is what matters. My three-pillar framework for AI transformation explains why investing in just one pillar produces a pilot that plateaus.
The Future of Data Science in an AI World
AI didn't kill data science. It killed the version of data science that was just running queries and writing decks. The work moves up the stack - from analysis to system, from model to harness.
From ML Models to AI Analytics Systems: What Transfers, and What Doesn't
ML platform discipline transfers to AI analytics systems - feature stores become semantic layers, drift detection still applies. The serving layer is where the bar goes up, and eval is the piece without a shared standard.
The 80% Is Commoditized
Anyone can build now. The next decade belongs to whoever closes the last 20% with real memory, context, and precision.
Nobody's Hiring for Output Anymore
The labor market isn't disappearing. It's repricing around who can think, decide, and own outcomes.
We're in the "Agents as Employees" Phase. It Won't Last
Most teams still define agents as human roles. The longer-term shift is toward workflow-native systems, review discipline, and context infrastructure.
AI-Native Is Not AI-Assisted
There's a difference between using AI and restructuring around it.
The real value of AI in analytics isn’t just answering questions. It’s closing the decision loop.
Decision latency is the real metric, not chat interfaces or dashboards.
OpenClaw Week Two: Four Observations on AI-Native Systems
Capability is no longer the bottleneck. Operational economics, context engineering, and governance are.
The Skills That Survive AI
What separates people who thrive in AI-native orgs from those who are stuck.
AI at Scale Isn't a Model Problem. It's a Systems Problem
Most companies don't have an AI problem. They have a data readiness and decision problem.
What AI-Native Actually Feels Like
Memory, tools and reusable workflows turn an AI assistant into an operating layer. The enterprise gap is the environment around the model.
What the Knicks' 2026 Title Teaches Us About Analytics
The Knicks' championship run is a reminder to distinguish player impact, team forecasts and the decisions made inside a series. An unexpected result is a reason to investigate, not proof that a model failed.