About
A small AI shop in Cleveland, Ohio — and not new to this.
Today Thalatta is mostly one set of hands — mine, Jeff Tirey — with a trusted collaborator or two when a job calls for it. That means you work directly with the person who designs and builds your agent, not a layer of account managers between you and the work.
And I'm not new to AI. I've spent two decades building software, and I was working in machine learning — prediction, decision trees, the algorithmic guts of it — as far back as 2010, long before LLMs turned “AI” into a household word. What's new is the tooling, not the discipline.
Want to know more about me? Sign in to Glass and ask the agent — it knows my background.
What we've built
Agents already doing real jobs.
A sample of what we've shipped — each one pointed at a single, real job:
RFP agents
Draft first-pass proposal responses from approved language — and have won new business.
HR agents
Answer the routine questions on repeat, giving the team back its hours.
Event agents
Move knowledge quickly through a conference, to whoever needs it next.
Interview agents
Train sales teams with realistic practice, on demand.
SOP agents
Apply a company's standards to work before it goes out the door.
Ad-approval agents
Watch brand safety on creative, every time.
Why “Thalatta”
Thalatta is the Greek word for the sea. A company's memory should work like one.
The sea is the reservoir every river flows into — deep, persistent, and nothing of value is lost in it. That's what a knowledge layer is for a business: somewhere your hard-won context accumulates and stays, ready the moment an agent needs to know what your company knows.
But a sea is vast, and that's the catch. Memory only earns its keep if you can find the right thing at the right moment — and drawing the one fact a job needs out of everything a company knows takes skill. That's our work, and why the mark is an amphora: a vessel made to hold the sea and pour exactly what's wanted.
How we work
We join your team — and we put a number on it.
This isn't built at arm's length. We come into your building and work alongside your people, because the only way to automate a job well is to understand how it really gets done. That's also how the relationship starts — and the relationship is the point. Agents earn their keep over months and years, so we build for the long haul, not a one-and-done handoff.
We like winning, and here winning is measurable. Before a line of code, we agree on what a good return looks like — hours saved, work shipped, business won — and we put a number on it. None of this is fluff: every agent has a cost and a value, and we keep both in plain view. The goal is never a clever demo; it's a system that pays for itself, and that you can put your name on.