Agentic Engineering

Software delivery with coding agents as part of the engineering system. Harnesses, guardrails, and quality gates that keep agent-written code reviewable, and telemetry that shows what the agents cost and what they produced.

Agent harnessesGuardrailsQuality gatesAgent telemetryDurable workflows

What it covers

Plugins, hooks, skills, and subagents for Claude Code, Codex, and Oh My Pi

Repository guardrails and CI quality gates that agents cannot bypass

Telemetry for agent sessions: tokens, tool calls, failures, and cost

Durable agent workflows with typed outputs and idempotent side effects

Least-privilege, read-only sandboxes for agents working on untrusted input

Typical work

Agent harness setup for a team's repositories: instructions, skills, and guardrails

CI quality gates and static-analysis rules for agent-written code

Agent session telemetry: tokens, tool calls, failures, and cost

Agent workflows for triage, investigation, and pull request review

How I think about it

Agents make code cheap to produce and expensive to review. The leverage is in the system around them: clear instructions, checks that fail loudly, and telemetry that shows where the time and tokens went. I build that system so the output stays something a team can trust.

Other expertise

Automation & Applied AI

Manual work mapped and measured, then automated where it pays off. Models chosen on evals and cost, with guardrails.

Platform & Infrastructure

AWS and GCP platforms defined in code. Kubernetes, GitOps delivery, observability, and cost control.

Security & Compliance

Identity, least privilege, audit trails, and data boundaries designed in. Architecture that holds up under HIPAA, SOC 2, and ISO 27001 review.

Data Engineering

Ingestion, enrichment, and event pipelines. Data modeled and accessible for products, analysis, and automation.

Product Engineering

Web, mobile, and backend products from first version through launch and iteration. Full stack, from data model to interface.