Coaching
Sixty focused minutes on your codebase, in your tools. Walk away with a working AI workflow and the files to keep using it — not a reading list.
See coaching →A Training Pack is the self-serve on-ramp. When you'd rather have the operator in the room — fixing your stack, teaching your group, or steering your team — there are three higher-touch doors. Same person who ships games and AI products for a living, working on your actual tools.
Sixty focused minutes on your codebase, in your tools. Walk away with a working AI workflow and the files to keep using it — not a reading list.
See coaching →The operator in your room, teaching your whole group hands-on — on the tools they already use, or the PickBits University curriculum taught live instead of read alone.
See workshops →Multi-week engagements that embed AI workflows into how your IT or ops team actually works — 2–5× output by automating the busywork.
See consulting →You're under pressure to deliver more with the same headcount. Status reports, alert queues, manual documentation, repetitive tickets — your team's time is consumed by work AI can handle. We identify, build, and embed the workflows that cut that overhead, so your people focus on what actually moves the needle.
We find the workflows where automation delivers meaningful ROI — and honestly flag the ones where it won't.
Working AI workflows wired into your real tools and data — not slideware. Shipped, tested, in production.
Your team learns to run and extend what we built. Prompt vaults, configs, and playbooks they own.
Track hours recovered and cost shifted, then roll the pattern to the next function.
A 60-minute 1:1 on your stack — building prompts, configuring tools, working through real tasks. Every Training Pack includes one session voucher to kick things off; buy more standalone any time.
// every session produces concrete, reusable outputs
A curated set of prompts and templates built for your role, your stack, and the tasks you do most. Yours to keep and extend.
# PR Review — Rails API
You are reviewing a pull request for a Rails 7 API.
Focus on:
- N+1 queries (check includes/preload)
- Missing index migrations
- Serializer field exposure (no internal IDs)
- RSpec coverage for new controller actions
Format as:
## Summary (1–2 sentences)
## Issues (blocking / non-blocking)
## Nits (style, naming)
Copilot, Cursor, Claude — whatever you run. Optimized settings, integration patterns, and the rules files that make them behave on your codebase.
# Cursor Rules — React + TypeScript
You are an expert in React 18, TypeScript, Tailwind.
Conventions:
- Functional components with hooks only
- Prefer server components (Next.js 14)
- Zod for runtime validation at API boundaries
- Co-locate tests: Button.tsx → Button.test.tsx
Never:
- Use `any` — use `unknown` and narrow
- Leave console.log in committed code
A step-by-step plan documenting everything we built — clear enough to follow without us, shareable with your team.
# Playbook — Maria, Senior DevOps
# Stack: Terraform + AWS + GitHub Actions
## What We Built
1. IaC Review Prompt — catches security groups,
missing tags, drift risk before apply
2. CI Pipeline Generator — service name in,
deploy workflow w/ staging gate out
3. Incident Runbook Writer — post-mortems
into on-call runbooks
## Action Items (This Week)
☑ Add IaC review prompt to shared vault
☑ Test pipeline gen on billing-service
☐ Share runbook writer with on-call lead
Questions come up as you apply the workflow. Reply to your follow-up email anytime that week — responses within 24 hours.
_ref_id?Exception: fields matching *_ref_id are internal reference keys, safe to expose. Updated file attached — also patched your debug prompt with the same list..rubocop.yml to the context window. Here's the updated prompt that pulls it in automatically — drop it in the same directory.// from booking to follow-up
A 10-minute form on your role, stack, and goals so we hit the ground running.
Screen-share on your real codebase — building prompts, configuring tools, real tasks.
Prompt library, tool configs, and a written playbook — packaged and ready to use.
Reply to your follow-up email as questions come up; responses within 24 hours.
Coaching levels up one person; consulting is us building it into your team. A workshop is the operator in your room, teaching your whole group hands-on — on the AI tools they already run, or the PickBits University curriculum delivered live as a class instead of read alone. On-site or remote, scoped to your stack and your day.
A hands-on session on the tools your group actually uses — Claude Code, VS Code, GitHub Copilot, Claude, ChatGPT — worked through on your real tasks, not toy demos.
Any PickBits University track — Vibe Coder, Field Guides, Vibe Management, Solo Business — delivered as a live class, with Q&A and your team's real work done in the room.
We come to you, or run it live over video for a distributed team. Half-day or full-day formats, sized to the group.
The group walks out with the prompt vaults, configs, and playbooks from the session — yours to keep and extend after we leave.
When the goal isn't a one-off workshop but a whole organisation learning to design, ship, secure, and operate AI agents — the four-track PickBits University curriculum delivered as a structured program. Self-serve for the individual contributors, live workshops for the team, embedded engagements where the workflows actually need to ship. Built in Phoenix, taught worldwide.
Vibe Coder (builders), Vibe Management (IT teams in production), Field Guides (knowledge workers), Solo Business (operators) — each track its own mental model and working stack.
Self-serve Track Packs for individual contributors, live workshops for the whole department, embedded consulting where the workflows need to ship into your real stack — same curriculum, three velocities.
A themed 60-minute workshop your team can attend free, launching soon. Use it as a sampler before you scope the bigger engagement, or as ongoing reinforcement after. Join the waitlist to get the invite the day it opens.
Tool-agnostic across Claude, ChatGPT, Cursor, GitHub Copilot, Microsoft 365 Copilot, MCP. Taught by the same person who ships agents into production every week — no junior associates, no cookie-cutter slides.
A live bilingual storefront, built end-to-end and shipped to production.
A full retail platform from scratch — bilingual EN/ES catalog, appointment booking, JWT-secured admin back office, and PostgreSQL-backed frame inventory. Live at excellens.us.
Read the full case study →Coaching vs. workshop vs. consulting, pricing, format — answered on the Common questions page.
One hour on your codebase, a workshop for your whole group, or a multi-week engagement for your team. Each starts with a real conversation.