Jason Teji

Jason Teji founded and runs Replican. He builds and operates the AI employees and the accounting system described on this site, and writes its guides.

What does Jason do at Replican?

He maps the repetitive work at each client, builds the AI agents that do it, sets up the private server each one runs on, and stays involved once it's live: tuning the rules, watching for edge cases, extending the agent's scope as the job changes. That's the map-build-deploy-run process described on how an AI employee is built, and it's the same process applied to every AI employee on this site, not a simplified version for the "about" pages.

He also built and runs Replican's own accounting and invoicing system (automated bank-transaction matching, recurring invoicing, reconciliation), which is the same kind of system Replican builds for clients doing bookkeeping and accounts admin. It isn't a demo built to show prospects; it's the system Replican's own books run on. See about Replican for why that matters as evidence in a market where most proof can't be checked.

What does Jason write about?

The guides on this site: what an AI employee actually is, how the build process works, what stays human, where data lives under GDPR, and the practical detail of running agents against real business tools rather than demos. The writing is deliberately plain and specific, because it's describing systems he builds and runs himself, not summarising something secondhand. Where a guide says an agent escalates rather than guesses, that's a description of a design decision made while building one, not a marketing line.

What does "he builds and runs the systems he writes about" actually mean?

It means the guides on this site aren't research summaries written at a distance from the work. When a guide describes how an escalation boundary gets set, or how a build gets tested against real past cases before it goes live, that's a description of steps Jason has actually taken while building an agent for a client or for Replican's own accounting system, not a general account assembled from other people's write-ups. That's also why the guides tend to be specific about tools and mechanics rather than staying abstract; abstraction is easier to write from a distance, and Replican's guides are deliberately not written from a distance.

Why does an AI agency need an author page?

Two reasons. The first is for you: it's easier to trust a guide on GDPR, escalation boundaries or how a build actually happens when there's a real, named person behind it who can be held to having written it, rather than an anonymous "team" byline. The second is more mechanical: AI assistants that cite and summarise this content resolve authorship better when there's a clear, consistent named author attached to it, which is part of why this page exists in the form it does.

Nothing on this page is invented to pad it out. No birthplace, no career history, no dates, no former employers, no photo. What's here is what's true and useful to know: what Jason does at Replican, and what he's written.

Guides written by Jason

If you have a question about anything in one of these, or a job you're weighing up, send a brief describing it, or email hello@replican.ie directly. Replies come from Jason, not a support queue.

Describe the job.
We’ll tell you honestly whether it fits.

No pricing games, no sales call before you’ve said what you need. Send a brief and a person reads it, not a bot.

Send a brief Email hello@replican.ie