2025Platform
AI audit review platform
Architected an audit review platform with Claude, MCP servers, LangChain and retrieval for a large consumer bank.
COACHPrincipalCoach◉ Charlotte, NC
Vick K.✓
"A single great agent is a parlor trick. A team of them, with a command center, is a company."
Get matched with the right coach
Senior full-stack engineer with ten-plus years at banks, Rivian and healthtech. MS in Computer Science (Machine Learning) from Stanford. Runs agent pods in production.
Request a coach →Tell us the problem. We match you with the coach who fits it within 48 hours. Independent, not affiliated with Anthropic.
YesVirtual 1:1
coaching
coaching
9Engagements
2Years with Claude
1Workshop hubs
·CCCA credentialed
About
A coach who respects your time.
I have spent more than ten years shipping software in places where it has to work: account-opening flows at a regional bank, a microfrontend standard adopted across most of Rivian's applications, and accessible platforms at Bank of America and Ally. I hold an MS in Computer Science with a focus on machine learning from Stanford.
I architected an AI audit platform with Claude, MCP servers and retrieval that cut the review of about sixty audits per cycle from weeks to two days at most. On my own projects I run pods of Claude agents and review their output before anything reaches a human reviewer.
I coach developers and tech leads on agent team design, evals, test coverage and the day-to-day habits that keep agent-written code honest.
Best fit
Team1-50 people
StageSeed to enterprise
IndustriesFintech and banking, healthtech, mobility, B2B SaaS
RolesEngineers, tech leads, founders
Not forPure content marketing, demoware
Methodology
Three principles on every engagement.
i.
Design the agent team
Roles, handoffs and stopping conditions before any agent ships. The org chart for your agents matters as much as the prompts.
ii.
Tests and evals as the contract
Every agent has a written eval rubric and the code it touches has real coverage. No green checks, no production.
iii.
Review before it reaches a human
A first-pass review layer for agent output, so your senior engineers only see work worth their time.
Specialties
Toolkit
Tools Vick reaches for.
Primary model
Claude Opus 5.5 + Sonnet 5.5
Opus for planning agents, Sonnet for execution agents.
Editor
Claude Code
Where most of the engineering coaching happens.
Integrations
MCP servers
Connecting agents to real systems with clear permissions.
Want a coach like this?
Tell us the bottleneck in a short request. A person on our team reads it and matches you with the coach who fits it, over video or in person, usually within 48 hours.