AI dev studio & consulting · Switzerland, est. 2026

We run AI inproduction.

Most teams are still running demos.

AI in production.Your agents work Monday and fail Wednesday, and the AI bill keeps climbing. tsukumo turns your dev team into agentic operators, devs who run fleets of AI agents in production, not a copilot that finishes their lines. Prod-grade, on the stack and standards you already have. We build this for ourselves; we'll do it with your team.

Talk to us about your team See what we've shipped

We build WRAI.TH, trovex and yoru, and run agentic development in production with Clareo Systems and a Swiss real-estate group.

We don't sell you a model. We change how your team operates.

The gap between a copilot in the editor and agents that actually run work in production, that's where most teams are stuck.

01
You bought the seats. Your devs have Claude Code. It autocompletes, it answers, it's a better search.
02
You can tell that's maybe 10% of what this is, you keep seeing demos of AI doing far more.
03
None of it survives contact with your real codebase, your real standards, your real review culture.
04
It isn't a license problem or a model problem. It's an operating problem.

How do you make agents reliable, cheap, and trustworthy on a real codebase, without breaking what works or scaring the people who built it. That gap is what we do.

We build our own production tooling.

The software we ship for ourselves is the proof: we run our own agent fleets to build it. Self-proof, we do to ourselves what we'd do for you.

orchestration

WRAI.TH

Runs and coordinates fleets of coding agents over a relay.

context

trovex

Feeds agents the right context at the right time.

~60% fewer tokens

observability

yoru

Tells us what the agents actually did, in production.

We don't run on testimonials we don't have.

Here's what's real, in plain words. We don't publish a number we can't stand behind, so each case shows the honest outcome; verified figures get added as they land.

scrollfour engagements

Consulting · quant finance

Clareo Systems, agentic-development training for a quant team.

Clareo Systems: agentic-dev training for a quant team

Their own people running agents, safely, inside a correctness-critical environment.

Context: Clareo Systems is a quantitative finance firm. Correctness and risk control are non-negotiable, and the usual AI sales pitch, faster and more output, points the wrong way in a place where a confidently-wrong change is the expensive failure mode. The same properties that make agents valuable, speed and volume, are the ones a risk team is right to distrust.
The gap: A strong team using AI as a copilot, with the ceiling that implies: autocomplete in the editor, a human typing every consequential line. The jump to operating agents on real work is a skill and a process change, not a license, and in a controlled environment you can't bolt agents on top of the controls. They have to run inside them.
What we did: Agentic-development training for the team, then the agentic process and guardrails built for their environment, not a generic playbook: a human gate on anything irreversible, scoped permissions, and review at the right altitude. We upskilled the CTO to lead it, because the goal was their people operating agents on their standards, not a dependency on us.

Outcome

Their CTO and team now run agentic workflows in their own repo.

Read the case →

Studio · real estate

Agentic marketing and platform for a real-estate group

Marketing and product, both running on agents.

Context: A real-estate business that needed to scale content and marketing output (market notes, area guides, listings, editorial) and modernize its platform, without scaling headcount to match.
The gap: Manual content and marketing couldn't keep pace, and the platform wasn't built for an agent-first way of working.
What we did: Automated their acquisition marketing, stood up agentic blog-content creation in their voice, built an agentic-first platform, and ran GEO/SEO so the content gets found, including by AI engines.

Outcome

An agentic-first platform producing content and acquisition continuously.

Read the case →

Self-proof · own venture

Our own studio running internal AI and autonomous dev teams.

Our own studio: agentic operations in production

We run what we sell, on our own operations and products.

Context: We are our own first client. We advise on running AI in production, and we run it on our own operations and products, which is the real test of everything we sell.
The gap: Real business operations (accounting, billing, payroll) and product development that a small team can't carry at the pace we hold ourselves to, without agents doing real work.
What we did: Built internal AI for accountants, an automated billing pipeline, and a payroll engine, and stood up autonomous dev teams that build and ship. The operating model we transition clients into is the one running our own business.

Outcome

AI-run accounting, billing, and payroll, plus autonomous dev teams shipping product.

Read the case →

Studio · civic tech

Open Truth, an algorithm plus agents for political-claim verification.

Open Truth: agentic verification for political claims

Applying agents to a problem most teams wouldn't attempt.

Context: Open Truth tackles verification of political claims, a domain where being right, sourced, and unbiased matters more than being fast or clever.
The gap: Claim verification at scale is beyond manual fact-checking, and a naive LLM that just answers is the wrong tool: it has to be sourced, checkable, and resistant to confidently-wrong output.
What we did: Built the algorithm plus the agent system that verifies political claims, designed so the output is grounded and auditable rather than an opinion. A novel AI system, not an off-the-shelf chatbot.

Outcome

An algorithm-plus-agents claim-verification engine running in production.

Read the case →

Honesty rule: no invented metrics, logos, or quotes, ever. Until each case has a real, verified outcome, it ships as an engagement summary without a fabricated number.

Two doors into the same capability.

leg A01 / 02

Studio

We build it.

Product and AI systems, shipped to production at studio pace. The software we ship for ourselves (WRAI.TH, trovex, yoru) is the proof: we run our own agent fleets to build them. You get a product and a demonstration of the way of working.

Product builds
Agent fleets in prod
Owned outcomes

leg B02 / 02

Consulting

We transition your team.

We take a team that uses AI as a copilot and turn it into a team that runs agents as operators, in production. On your existing environment, your standards, your stack. We augment your devs; we do not replace them. The craft stays theirs, and the output multiplies.

Strategy
Implementation
Training

We augment your developers. We never replace them.

Your devs become the operators. The craft stays theirs; the output multiplies. That is the honest position, and the one that actually works.

Who this is for.

For

CTOs at startups and scale-ups who've felt the gap.
Teams who want to ship at a level seats alone won't reach.
Leaders who respect their developers enough to want them stronger, not sidelined.
People who treat AI in production as an engineering problem.

Not for

Anyone who thinks AI is easy.
Anyone who believes buying licenses is the same as capability.
Buyers who think the point of AI is to ship the same thing cheaper.
Teams looking for a demo to screenshot, not software to run.

This isn't about cheaper. It's about prod-grade output and a far more capable team than the one you already trust.

Start with the free check.

Two minutes, no login. See where your team sits between AI copilot and operator, and the one gap to close first. You keep the read; there's no sales call to get it.

Take the readiness check

Your devs are readyto be operators.Your tools arewaiting to wake up.

That's the tsukumo idea. Tell us what you ship and where AI keeps stalling, let's talk about your team.

or email hello@tsukumo.ch

We run AI inproduction.

Most teams are still running demos.

Talk to us about your team See what we've shipped

We build WRAI.TH, trovex and yoru, and run agentic development in production with Clareo Systems and a Swiss real-estate group.

The gap between a copilot in the editor and agents that actually run work in production, that's where most teams are stuck.

You bought the seats. Your devs have Claude Code. It autocompletes, it answers, it's a better search.

You can tell that's maybe 10% of what this is, you keep seeing demos of AI doing far more.

None of it survives contact with your real codebase, your real standards, your real review culture.

It isn't a license problem or a model problem. It's an operating problem.

How do you make agents reliable, cheap, and trustworthy on a real codebase, without breaking what works or scaring the people who built it. That gap is what we do.

Two doors into the same capability.

leg A01 / 02

Studio

We build it.

Product builds
Agent fleets in prod
Owned outcomes

leg B02 / 02

Consulting

We transition your team.

Strategy
Implementation
Training

Who this is for.

For

CTOs at startups and scale-ups who've felt the gap.
Teams who want to ship at a level seats alone won't reach.
Leaders who respect their developers enough to want them stronger, not sidelined.
People who treat AI in production as an engineering problem.

Not for

Anyone who thinks AI is easy.
Anyone who believes buying licenses is the same as capability.
Buyers who think the point of AI is to ship the same thing cheaper.
Teams looking for a demo to screenshot, not software to run.

This isn't about cheaper. It's about prod-grade output and a far more capable team than the one you already trust.

Your devs are readyto be operators.Your tools arewaiting to wake up.

That's the tsukumo idea. Tell us what you ship and where AI keeps stalling, let's talk about your team.

or email hello@tsukumo.ch