Skip to content
Done-for-you · Async only · No calls

When a product is not enough, I do the work myself.

Two engagements I run personally. Every inquiry goes to my inbox with a 24-hour response window. Async only. No calls, no calendar links, no discovery meetings.

Fixed price before work starts. First working output inside 48 hours. You own everything, no lock-in.

Two engagements. One clear brief each.

Audit is fixed price, five business days. Build is scoped per project with a fixed price once we agree on scope.

Prompt Audit
$499one-time · 5 business days
You send transcripts, I return a written action plan.
Top 5 patterns costing the most tokens or review time
Specific prompts to add as skills or slash commands
A rewrite of your CLAUDE.md with explanations
3-5 before/after examples from your own transcripts
A written walkthrough video (~10 min, async)
One round of written followup questions, no time limit
BEST VALUE
Custom Agent Build
From $5K1-3 weeks
I design and ship a Claude Code agent end-to-end for a specific workflow.
Working agent (skill files + prompts + runner script)
Evaluation harness with test cases your team can rerun
CLAUDE.md plus docs explaining how to extend it
Weekly async progress updates via loom and written report
Two weeks of written tweak requests after handover
Fixed price once scope is agreed

Services questions, direct answers.

Why no calls?
Async only. Every engagement runs on written briefs, written updates, and async walkthrough videos. It keeps both sides honest, timezone-independent, and forces the work to be legible. If you want a synchronous conversation, buy a product first and reply to the delivery email.
What if my situation is more complex than $499 audit but smaller than $5K build?
Reply to the audit inquiry with details. I will scope it as an audit-plus or route you to a targeted product (Skills Library, Cheat Sheet, a specific Pack) that covers your situation for a fraction of the cost. Honest sizing beats padded scope every time.
Do you offer refunds?
Digital deliverables are all sales final. If the audit's written output does not land, reply to the delivery email and I will rework the specific section that missed. For the build, the scope is fixed before work starts, and you see real output in the first 48 hours so course corrections happen early.
What kind of work do you take on?
PR reviewers, migration agents, domain-specific bug triage, customer-facing copilots, document ingestion pipelines, evaluation harnesses. If it involves Claude Code plus a specific workflow, I probably have shipped something similar before. Send a paragraph and I will tell you fit inside 24 hours.
Do you sign NDAs?
Yes, mutual NDA on request. Standard mutual template, no bespoke redlining. I do not name clients publicly without written approval.