Real results from AI: built and shipped
Most companies have people using AI. They just have to move from use to knowing how to achieve outcomes such as increased revenue, productivity, quality, customer satisfaction. Atelier at PurpleBlossom.ai closes that gap for mid-market companies. We put AI into real work and help you prove it paid off. Two lines of work, one practice.
Why this practice is different
We ship.Working software, built with AI, shipped with documentation and tests. Your team walks out with a setup tuned to your real work.
We tell you where AI doesn't belong.That honesty is part of the relationship we build with you.
Outcomes over adoption.We help you focus on results that you care about. Many companies are still measuring usage, which isn't the point.
Senior, three ways: build it, sell it, run it.Run by someone who ships production Claude Code work, has led go-to-market, has run the business, and can speak comfortably with your exec team. (See "Who's running it.")
Two lines under one roof
For engineering teams
Done-with-you adoption that gets your developers shipping with Claude Code deliberately. We can also design and implement a custom MCP server, or teach your team to build and assess their own.
Ship with Claude$15–22k
1–2 week adoption: Claude Code installed and tuned across the team, CLAUDE.md authored against your repo, a custom MCP server built against your real systems, pairing on live tickets, runbooks left behind.
Shipping with Claude Code (cohort)$30k
5-week team training for teams still building the habit. Structured, hands-on, with your codebase.
Custom MCP server build$10–25k
A working MCP server against your internal system, your stack, your auth. Flat fee per build.
AI tooling audit$5–8k
One-week read of how your team should be using AI, with a written report. The easiest way in.
Office hours retainer$1.5–3k/mo
A weekly call plus async access. Ongoing support for a team that's adopted and wants a senior hand nearby.
For business teams
Getting AI into the everyday work of sales, marketing, operations, support, and finance, with the judgment and verification step that keeps a wrong output from reaching a customer.
Executive Briefing$5k / $9.5k
A 90-minute working session (or half-day) that gives your leadership team a defensible answer to "where do we deploy AI first, and how do we govern it?" Leave with a written First Deployment Recommendation and a governance starter.
AI Outcomes Assessment$15k
A ~2-week fixed-fee diagnostic for teams that have already been doing AI and need to know if it's working. We locate you on the AI Outcomes Maturity Model per goal, define the metric and the measurement, and hand back a scorecard and roadmap. Vendor-neutral. We're not selling you any AI stack. Optional re-run at 6 months for $7.5k.
Deliver with ClaudeScoped per engagement
Done-with-you acceleration that takes a real business objective from coursework to results.
Where to start
| If you're… | Start with |
|---|
| An engineering team that's decided to adopt Claude Code | Ship with Claude |
| An engineering team still on the fence | AI tooling audit or cohort |
| A leader unsure where AI fits | Executive Briefing |
| A team a year into AI with no proof it worked | AI Outcomes Assessment |
| Not sure | A 15-minute intro call |
How we price
Fixed-scope, locked in writing before work begins. Single invoice, Net-15. No hourly surprises. Out-of-scope items are priced separately, never absorbed.
Productized engagements start at $15k; the session SKUs (Briefing) are priced below that as single-session products.
Reference discounts available in exchange for an on-the-record quote.
Who's running it
Brian Fromme is a senior software developer who ships with Claude Code, backed by two decades at Canonical and SUSE across engineering and go-to-market, plus a decade running his own consulting practice.
He builds AI that produces results and tells you where it won't help. Based in Colorado; works with teams across North America and beyond.
Next step
Book a 15-minute intro call.
A working conversation about your goals and whether one of these engagements is the right next move.
brian@purpleblossom.ai
The position, in one line
AI maturity isn't how much AI you adopt. It's AI dependably producing the results you set out to get.