What we build
Systems, not chatbots
Every build starts from one question: where does your team spend hours moving information between tools by hand?
AI agents
Agents that handle multi-step work: intake, triage, follow-ups, first drafts, and alerts, with human approval built in.
Knowledge assistants
Search and answers grounded in your own documents, with the source cited so your team can check it.
Connected data
Your practice, project, accounting, and CRM tools joined into one layer so the numbers agree.
Dashboards
One view of the work and the money, updated as it happens instead of rebuilt by hand each week.
Who we work with
Organizations that run on documents, deadlines, and contracts
Law firms
- Client intake and conflict-check triage
- First-pass contract and document review
- Search across past matters and briefs
- Deadline and renewal tracking
Construction and trades
- Job cost: budget vs actual, in one view
- Change orders that never get billed
- Pay apps, retainage, and contract dates
Professional services
- Accounting, engineering, and consulting firms
- Proposal and report first drafts
- Time, billing, and utilization reconciled
Nonprofits and organizations
- Grant and program reporting
- Member and donor requests routed and answered
- Board-ready summaries from scattered data
Different industry? If your team copies information between systems every week, it's worth a conversation.
How it works
Start with one workflow. Expand from there.
Audit
We map your tools and the manual steps between them, measure the hours, and pick the first workflow worth automating.
Build
A working system on your real data, tested with the people who will use it before it goes live.
Run and grow
A monthly retainer keeps it accurate as your work changes, and adds agents as new needs show up.
How we build
Built for places where mistakes are expensive
A person approves
Agents draft and flag. Anything that leaves your organization or changes a record waits for a human.
Every step is logged
You can see what an agent read, what it did, and why, after the fact.
Your data stays yours
Systems run in accounts you own, with access limited to what each workflow needs.
Who you work with
Sarian Potin, founder
AI Systems Architect
I'm a software engineer with 7+ years building production systems, much of it in regulated environments: healthcare, finance, and legal. I've built a production document retrieval system for a law firm, and I've done model evaluation and responsible AI work.
I started Seccho AI to bring that level of engineering to small and mid-sized organizations. Seccho AI is new, so you work directly with me from the first call to the finished system.
- Focus
- Agentic AI systems and data
- Industries
- Legal, healthcare, finance, enterprise
- Education
- B.S. Computer Engineering
- Based in
- Minneapolis, MN