Inspired By FrustrationFractional AI CTO. Senior AI Systems Architect.
Ships agent fleets in production. 100 PRs/day with one human. Twelve years on the operating side of software, now an Amsterdam-to-Atlanta studio that finishes the demos other shops abandon at the wall.
Inspired by frustration — I mean that literally. Every product with my name on it started as something that annoyed me enough to fix. The studio is named after the feeling for a reason.
// the operator
- based
- amsterdam → atlanta
- operating
- 12 years
- studio
- Inspired By Frustration
- shape
- one operator, one swarm
The studio is named after the feeling. I mean that literally.
// the origin · “i could never finish the lovable app”
The demo is the easy part. Orchestrating agents to ship the rest — that's the job.
I watched no-code and AI demos die at the same wall — eighty percent done, never shipped. The frustration that named the studio became the orchestration layer that gets them across. AppHandoff conducts a swarm of agents: contract-aware, gated, shipped as production code.
That's the craft — not prompts, not prototypes. Agent orchestration, the CI fleet that keeps it honest, and the business judgment to aim it. The tooling that runs the swarm and the CI that catches it are my own products, running in production. I don't sell a method I haven't shipped.
Six rules. Every one of them has a receipt.
Lead with the human.
Business judgment picks the bet. The swarm is the engine, never the headline — and never an excuse.
Receipts, not claims.
Named products, real numbers, real dates. If I can't defend it in a sales call, it doesn't go on the page.
One operator owns the spine.
Not five vendors and a Gantt chart. One throat to choke, and every merge into main is mine.
Agents do the volume.
Up to thirty in parallel, coordinated by tooling I built. They write the volume; I review what's risky.
Governance first.
Branch protection, ship gates, evals, audit. The bar I sell to enterprises is the bar I run on myself.
No lock-in.
Your accounts, your code, runbooks included. Fire me and keep shipping the next morning.
// every frustration became a product
The name pays off. Four times so far.
Specifically. Every one of these is a real pattern.
Agencies that ship Jira tickets instead of working software. Velocity dashboards aren't a deliverable.
Consultancies that recommend but won't build. A 200-page deck is not the same as a thing that runs in production.
Vibe-coded MVPs with no architecture underneath. They demo well on a Tuesday and fall over on a Thursday.
Lovable prototypes treated like production systems. Lovable is fast; the backend, auth, MCP layer, SEO, and handoff discipline still need a senior owner.
AI demos that never ship. Looks great on LinkedIn, never makes it to a customer.
Juniors-plus-AI as a delivery model. AI is fastest in the hands of seniors who can tell when it's wrong. Without that, it's a defect factory with autocomplete.
Six-month timelines for things that should ship in weeks. The estimate isn't the work — it's the ceremony around the work.
If any of that sounds familiar, you already know why I started this.
Right fit. Wrong fit.
// who this is for
- Founders who need to ship now and don't have time for a six-week kickoff.
- Exec teams stuck mid-translation, where the engineering org and the business org keep talking past each other.
- Companies adopting AI tooling — Lovable, Cursor, Claude, MCP — that want the speed without the quality cliff.
- Lovable teams that need a top-ranked developer to connect generated frontends to production backends, APIs, and MCP workflows.
// who this isn't for
- Buyers who need a 200-page deck before any code exists.
- Engagements measured in headcount rather than outcomes.
- Projects that require a twelve-person ceremony before a button gets clicked.
If you're either of those, no offence — I'm just not the right shape.
I'll tell you what's worth building — and what isn't.
Bluntly. If an idea won't fly, you'll hear it from me before you fund it — not at the demo. The Dutch in me.