Nadia Okonkwo
Studio lead & chief scope assassinRuns the studio and the first call. Has removed roughly 400 features from roadmaps and lost no sleep over any of them.
About · the studio
Six people. One room. No account managers.
We started Bolt because we kept watching good ideas die in a discovery phase. Nine years later the rule is still the same: the people who scope your project are the people who build it.
How we got here
It started as a complaint.
In 2017 Nadia and Theo were both leaving agencies — different agencies, same complaint. Clients arrived with real problems and left with a slide deck, an invoice and a Gantt chart that had already stopped being true.
The first Bolt project was a booking tool for a friend's yoga studio, built over five weekends in exchange for a very generous number of classes. It's still running. Nobody has touched the code since 2019, which we consider the highest compliment a codebase can receive.
We've stayed small on purpose. Six is the number where everyone still knows every project, nobody needs a status meeting to find out what's happening, and we can turn down work without it threatening payroll. Growing would make us worse at the only thing we're good at.
2017
Two people and a yoga studio
Weekends, a borrowed desk, and a booking tool that still works.
2019
The eight-week rule
After one project sprawled to seven months, we started refusing scopes we couldn't finish in eight weeks.
2022
Fixed prices, published
We put our rates on the website. Sales calls got shorter and much more honest.
2026
Sixty-one products later
Still six of us. Still above the bakery. Still arguing about scope in week one.
01
The whole studio
Not "leadership". Not "a selection of our team". This is everyone — and every one of them will be on your project.
Theo Lindqvist
Principal engineer & resident pessimistFifteen years of backends. Asks "what happens when this fails at 3am?" in every architecture review, which is why it doesn't.
Priya Raghunathan
Design lead & empty-state enthusiastBelieves a product is judged by its worst screen, so she designs the error states first. Runs every user session we do.
Marcus Beaufort
Mobile engineer & offline evangelistSwift and Kotlin. Tests every build on the Lisbon metro because if it works underground it works anywhere.
Jules Aramburu
Growth engineer & funnel plumberHalf analyst, half front-end. Will not run an experiment until the tracking is clean, no matter how much anyone begs.
Wren Castellanos
Product designer & chief question askerJoined from a hardware company, which is why she keeps asking who's actually holding the phone and what else they're doing.
02
How we think
Four things we'd still believe if they cost us the project. They occasionally do.
Small scope, real launch
A shipped thing that does one job beats a beautiful thing that's still in staging. We'd rather hand you something narrow and alive than something broad and theoretical. You can always add. It's much harder to un-launch.
Opinions, offered early
You're not paying us to nod. If we think the pricing model is wrong or the second feature should be the first, you'll hear it in week one — clearly, once, with reasons. Then it's your call and we build what you decide.
Show, don't report
No RAG statuses, no burn-down charts, no eleven-page weeklies. A staging link that updates daily and a demo every Friday tells you more than any document, and takes us an hour instead of a day.
Leave it better than we found it
Readable code, a README someone can follow on their first morning, decisions written down with the reasoning attached. The measure of a good handover is that you stop needing us and don't resent us for it.
03
What we don't do
Saying this out loud saves everybody a fortnight. If your project needs one of these, we're genuinely happy to point you somewhere better.
- Staff augmentationDropping two contractors into your standup and billing hourly. We take on outcomes, not seats.
- Rebuilds with no reasonIf your stack is unfashionable but works, we'll tell you to keep it and spend the money on customers.
- Discovery phases that cost more than the buildSix weeks of workshops and a PDF is not a product. One call and a written brief will do.
- Blockchain-because-the-board-askedIf there's a real reason, great. If it's a slide in a fundraise, we'll pass.
- Fixed-scope RFPs with eleven biddersWe can't do our best work as the cheapest line in a spreadsheet, so we don't try.
- Gambling, payday lending, dark patternsNothing designed to work best when the user loses. Not a moral flex — we're just bad at it.
A normal week
Mondays are for arguing.
The week opens with an hour where anyone can challenge anything — a scope call, a component, a price we quoted. It's the only meeting nobody is allowed to skip, and it's why bad decisions rarely survive to Wednesday.
Fridays are demos. Every live project shows something that runs, to whoever's around, including clients who want to drop in. Nothing is presented from slides. If it can't be clicked, it isn't done.
Between those, it's quiet. No standups, no Slack-response expectations, and a hard rule that meetings need an agenda in writing or they get declined. Most of us write more than we talk.
We read everything ourselves
Think we'd get on?
The best projects start with a founder who's slightly annoyed about something specific. If that's you, we'd like to hear about it.