AI agents
RAG agent over the company's own knowledge
Crassula
A retrieval-based agent that answers from the company's own documents and data rather than from a general-purpose model.
European client
AI agent development for your processes
We take the process apart, build the agent for it and run it inside your own infrastructure: roles, spend limits, mandatory approvals and an audit trail of every action.
30 minutes, no deck, NDA if you need one. You describe the process; we work out how many people and hours it burns today, what can come off and what should stay manual. If it is not worth automating, we say so.
10 yearsshipping to production since 2016
20+products shipped for clients
3products of our own: AgentArea, AA.Docs, Constract
500+tool integrations in AgentArea
The team has worked with
Selected work
What we have built and for whom
Integrations, data and infrastructure are exactly where AI projects fall apart. We were doing that work long before anyone called it AI adoption.
Data infrastructure
Open-source ELT with 300+ connectors
Airbyte
A connector-heavy data platform used by hundreds of engineering teams, where a broken sync is visible the same day.
YC W20 · 300+ connectors
Distributed systems
Privacy-focused Layer 1 blockchain
Aleph Zero
Cryptography and consensus with sub-second finality, in a system where a mistake costs users money directly.
Sub-second finalityWhat people come to us with
Approvals live in email and chat
A contract travels through a chain of messages, and only the last person to reply knows where it stands.
The ERP and the CRM both exist, but data moves by hand
Someone opens two systems side by side and retypes rows. By evening one of them already holds different numbers.
It has to run inside your own perimeter
The cloud service passed every demo and stopped at the security review.
You tried AI already and it never left the pilot
The pilot was shown at a meeting, everyone nodded, and it never reached anyone's work.
We ask you to name an owner for the process on your side: someone who settles the open questions and signs off on each step.
What comes off your people
Three processes your company is certainly still doing by hand.
Not consulting, not development in general. One process, a fixed price, and a working system inside your own infrastructure. Then the next one.
Contract and procurement approvals
A route instead of an email thread: roles, deadlines, owners. The system checks the paperwork is complete, compares terms against your template and escalates only what deviates. You can see who is sitting on a document and for how long.
Inbound document handling
Invoices, delivery notes, specifications, supplier mail. Data lands in your system of record without manual entry, and discrepancies get flagged before signature rather than after.
Estimates, pricing and reporting
Scope and cost estimation, recurring reports and reconciliations — the work someone currently assembles by hand from three systems every Monday.
Our products
We don't describe this. We operate it.
The three processes above are three products of ours that already run. AA.Docs routes approvals and reads the documents, Constract prices the work, and AgentArea keeps control underneath: roles, limits, approvals, audit. Your project does not start from an empty repository.
Platform · open sourceAgentArea
The control layer for agent teams: roles, budget limits, approval rules, audit trails and 500+ tool integrations. Self-hosted, so the data never leaves your infrastructure.
Runs on AgentAreaAA.Docs
Document approval stops being an email thread. Contracts, invoices and specifications get a route, a deadline and an owner, while the agent reads the document, pulls out the data and flags the risks.
Runs on AgentAreaConstract
A renovation estimate in two minutes: floor plan in, materials, labour and a client-ready proposal out. Recalculates the moment the scope changes.
AgentArea is open source — read it before you talk to us: github.com/agentarea/agentarea
Running inside your perimeter
The questions a security team asks on the first call.
- The whole system sits on your servers.
- Rights are set per role. Agent actions carry spend limits, and expensive steps go to a person for approval.
- Every action is written to a log together with who approved it. The log lives on your servers too.
- Anything the agent finds doubtful goes into a queue for a human. You define what counts as doubtful.
- AgentArea is open source. We send the repository link before the call so your security team can read the code before any contract.
How a project runs
Four stages: from an intro call to a system in your own infrastructure
We take one process rather than the whole company. That is what makes a fixed price possible and lets you judge the result in two months instead of a year.
Intro call, 30 minutes
You walk us through the process as it works today. We tell you what an agent can take over and what it should not touch. Free.
Audit, 2 weeks
We map three to five processes and count the people, hours and money each one costs. You get the numbers, an implementation plan and a working prototype of the first.
Build, 6 to 8 weeks
The system goes onto your servers: integrations with your CRM, ERP and document flow, access rights, an audit trail, spend limits and approval on expensive operations.
Support and the next process
We watch quality in production, take on the next area, and train your team to run the system without us.
Agent University
We teach teams to build agents, not only build agents for them.
We run AI agent courses at leading universities and inside corporate programmes. The material is about how agents work in general — architecture, tools, memory, evaluation and the point where an agent stops being the right answer — not about our stack.
AI agents from prototype to production
Agent architecture, tool use and memory, orchestrating several agents, evaluation and observability. Every module ends in working code rather than slides.
Where agents pay off, and where they don't
How to size the economics of a process before automating it, which decisions an agent should never own, and how to brief a vendor and judge what comes back. For CTOs, COOs and process owners.
Ship your first agent, with us alongside
Your team takes one of your own processes to a working agent under our guidance. You end up with a live system and people who can build the next one without us.
An open Agent University platform is in the works. Join the waitlist · Talk about a corporate programme
What clients say
The thing clients mention first is that the team works things out on its own.
That matters most when you are letting an outside team into systems that are already live.
They had incredible responsiveness, a great understanding of what we want to do, and great autonomy.
John LafleurCOO, AirbyteJamakase Labs showcases deep expertise in web and mobile applications, successfully addressing any challenges.
Jonathan BenassayaCo-Founder of Deezer; now Senior Director of Product Management, MetaI'm impressed by their ability to meet those changes on very short notice while keeping with deadlines.
Andrew BondarenkoFounder & CEO, OculisaStraight answers
The demo takes a week. Getting security to sign off takes six months.
Can it run in a closed or air-gapped network?
Yes. The platform and the models run on your own servers with no outbound calls. Updates and model weights move in whatever way your security team approves.
Are we locked into your platform?
No. AgentArea is open source and runs in your own infrastructure, in your cloud or on-prem. The implementation code is yours. Your team or another vendor can maintain it: the platform is not our secret, our engineering around it is what you pay for.
Does our data leave our infrastructure?
Only if you want it to. The platform is self-hosted, and we can run open models inside your perimeter when sending data to an external provider is not an option.
Our data is messy and undocumented. Is that a blocker?
No, that is the normal starting point. Cleaning up what the agent needs, and only that, is part of the audit and the build.
We tried this already and ended up with a demo. What is different here?
Scope and control. One process instead of a platform for everything, and an agent that runs with explicit roles, spend limits, approvals and an audit trail rather than on trust.
Who maintains it afterwards?
Either we do, or your team does. Handover and training are part of the engagement, and teaching engineers is something we do at universities and in corporate programmes anyway.
What does it cost to start?
The first step is a free 30-minute call. After that comes a paid teardown at a fixed price, ending in a calculation, a plan and a prototype. We quote the build once the teardown shows the real scope, rather than guessing up front.
First step
Name one process. We'll tell you what it is costing you.
30 minutes, no deck and no commitment. We work out how many people and hours the process burns today, what can come off and what should stay with humans. If the honest answer is that automating it does not pay, that is the answer you get.