Now

The field is where the roadmap gets honest.

This page tracks the questions, systems, and independent products shaping my forward deployed practice right now.

Current topics

The build direction.

Forward deployed engineering

Building a practice around discovery, technical scoping, end-to-end delivery, production adoption, and reusable field learning.

Applied AI systems

Working on agent runtimes, context, memory, evals, control surfaces, and human review where models meet production software.

Operational product surfaces

Designing cloud, infrastructure, media, and internal systems where state, permissions, failure, and recovery need to stay visible.

Independent products

Taking small products through code, launch, distribution, measurement, support, and the decisions that only appear after release.

Working bets

  1. The distance between a demo and adoption is becoming a first-class engineering problem.
  2. The most useful engineers will move between field context, product decisions, code, and deployment without losing the thread.
  3. Independent products are a practical laboratory for ownership: discovery, building, distribution, support, and iteration stay connected.

Open questions

  1. Which field discoveries should become reusable product primitives, and which should remain deployment-specific?
  2. How should AI systems expose state, memory, tools, evals, and human control when they reach production?
  3. What is the smallest complete deployment that can create a real feedback loop?