I build agentic developer platforms that make engineering teams measurably faster.
Vendor-agnostic. Production-grade. Outcome-focused. I help engineering orgs turn AI coding tools into developer platforms their teams actually use.
The problem you actually have
Your team has Claude Code, Copilot, Cursor, Codex, and three internal AI hackathon winners. Adoption sits at 15%. There's no measurement, security is anxious, and velocity hasn't moved.
The tools work. What's missing is the platform around them: shared configurations, org-specific context, evaluations, guardrails, and the enablement that turns a license bill into a force multiplier.
That platform is what I build.
What I deliver
A typical engagement runs in three phases over 10–14 weeks. Scope and sequence flex to what your org needs.
Discovery & strategy
I audit how your teams actually use AI coding tools today (Claude Code, Codex, Cursor, Copilot, Windsurf, whatever you've licensed) and map adoption gaps, friction points, and security blockers. We agree on the metrics that prove the platform is working: cycle time, PR throughput, eval quality, developer satisfaction.
- Tool usage audit across teams
- Adoption + friction analysis
- Security and compliance review
- Success metrics and instrumentation plan
Platform implementation
I build the platform layer around your tools: standardized configurations, a shared skill library, internal MCP servers exposing org-specific context, subagent orchestration patterns tailored to your workflows, and the codebase-specific evaluation harnesses that catch regressions before they ship.
- Standardized configs + skill libraries
- Internal MCP servers for shared context
- Subagent and orchestration patterns
- Evaluation + regression harnesses
- Guardrails, secrets handling, audit logging
Enablement
A platform nobody uses is a sunk cost. I leave behind onboarding docs, in-house workshops, and runbooks so your team owns the system on day one. Adoption is part of the deliverable, not a hopeful side effect.
- Team onboarding documentation
- Live workshops (recorded for replay)
- Operating runbooks for ongoing evolution
- Internal champion enablement
Who this is for
Series B–D startups
Scaling engineering fast, rolled out three AI tools, can't tell which are working.
Mid-market enterprise
50–500 developers across FinServ, healthcare, or SaaS. AI mandate from the top, chaos on the ground.
Platform / DevEx teams
Already own developer tooling. Need a senior partner who has shipped this before, not a deck.
Federal agencies
AI adoption mandate, existing tool licenses, no easy path to hire the internal expertise.
Best fit: engineering orgs of 50–500 developers. Big enough to need the platform, small enough to act on it.
Why me
You have alternatives. Here's the honest tradeoff against each.
I'm the fast, deep, neutral option: a staff-level engineer who has shipped this pattern, working directly with your team for the weeks it takes to land it.
Federal engagements
Federal agencies have AI adoption mandates, existing tool licenses, and procurement rules that favor specialist consultants over full-time hires. I work with civilian and defense agencies on short-form pilots and platform builds that fit within standard vehicles.
Engagement model
Fixed-scope, fixed-fee. Scoped after a 30-minute conversation, agreed in writing before kickoff.
Audit, gap analysis, success metrics, and a written platform blueprint. Sometimes runs standalone for orgs that want a second opinion before committing.
Full implementation across discovery, platform, and enablement. Scope and fee shaped to org size, tool footprint, and compliance needs.
Optional follow-on for ongoing platform evolution: new tools, new patterns, eval maintenance, and quarterly adoption reviews.
Ready to talk?
A first conversation is 30 minutes. I want to understand what tools you've rolled out, where adoption stalled, and what shipping faster would unlock. If we're a fit, the next step is a discovery sprint.