Who is Vasiliy Uvarov?

Vasiliy Uvarov is an applied AI engineer in London and Chief Data & AI Officer at Altery, a UK Electronic Money Institution. He has 23+ years in technology, 18+ of them in banking and fintech, across engineering, product and business. AI is also his personal passion: he independently builds and ships products outside the day job.

Is he a hands-on builder or an executive?

Both. He personally builds agent services, retrieval systems and independent products. His executive role includes AI transformation across 7 departments. Product ownership, commercial judgment and leading teams are part of his experience alongside implementation; the work record distinguishes what he built from the programmes he led.

What can I inspect before getting in touch?

The MIT-licensed agent and skills library is public at https://github.com/vasilyu1983/AI-Agents-public. Cosmic Copilot and AEO Platform have live product sites. The library is a curated public slice, not the private services themselves. The work record explains the fraud detection engine and the governance retrieval capability; technical walkthroughs of private work are available on request, subject to confidentiality.

Vasiliy Uvarov · Applied AI engineer and Chief Data & AI OfficerI build applied AI that holds up in a regulated business.

Applied AI is my craft and my passion. I build Python agent services, cited retrieval and evaluation tools that have to work in production. As Chief Data & AI Officer at Altery, and previously its Chief Product Officer, I combine that with product, business and engineering leadership. Outside work, I ship my own products for web and iPhone.

Get in touch Read the brief

Seat: Chief Data & AI Officer, UK EMI
Experience: 23+ years in tech, 18+ in fintech
Location: London · UK Global Talent Visa, no sponsorship

Selected work · From decision to delivery

What I build. How I think.

Read the work record

Applied AI · Onboarding and financial crime

Faster onboarding. Fraud found in the logs.

I led agent-assisted KYB onboarding, which now saves hours on each business case. I also built a fraud and PII detection engine, which I run ad hoc on exported production sign-up and sign-in logs.

KYB, per case: 5+ hours saved
Fraud engine: 40+ detection rules

As at October 2026, the KYB flow has run 800+ checks and ordered 500+ documents; that result belongs to the programme team. I designed, built and reviewed the engine, with AI coding agents writing much of the code; its 293 tests include parallel-versus-single-pass equivalence checks.

Read the engineering account

Private systems · Technical walkthrough on request

Context & evaluation · Governance knowledge

An answer with evidence. Or an honest abstention.

I built a retrieval capability that uses approval-gated sources, returns cited evidence and abstains when the corpus is silent. Source scope and citation support are part of the engineering.

How the evidence is handled

Independent product · Cosmic Copilot

An idea, built for web and iPhone.

I built Cosmic Copilot across native iOS and web: an astrology product with subscription infrastructure and paying subscribers. It brings product decisions, interface design and engineering into the same hands.

Read the product story

Inspect for yourself

Explore the public agent and skills library Visit Cosmic Copilot

The library is a curated public slice; private client work is excluded.

Marks a claim that can be verified against a public record, without an introduction.

Product, business & leadership

The judgment behind the build.

My 23+ years in technology, 18+ of them in banking and fintech, span engineering, product ownership, business development and executive leadership. I have been responsible for choosing what to build, leading the teams that deliver it, and understanding the commercial result.

At Altery I was promoted from Chief Product Officer to Chief Data & AI Officer. Today my AI work sits alongside organisational responsibility across 7 departments. Personal implementation and leadership are both part of the record.

Explore the career record

Product · ECOMMPAY · 2022–23: Led KYB onboarding changes: 90 days to 12.
Business · Moscow Exchange · 2021–22: Replaced 50% of legacy dependencies, unlocking £1.7 million a year.
Leadership · Bank Otkritie · 2018–21: Oversaw 230+ engineers across 21 teams.

About the person

A few useful answers.

Who is Vasiliy Uvarov?: Vasiliy Uvarov is an applied AI engineer in London and Chief Data & AI Officer at Altery, a UK Electronic Money Institution. He has 23+ years in technology, 18+ of them in banking and fintech, across engineering, product and business. AI is also his personal passion: he independently builds and ships products outside the day job.
Is he a hands-on builder or an executive?: Both. He personally builds agent services, retrieval systems and independent products. His executive role includes AI transformation across 7 departments. Product ownership, commercial judgment and leading teams are part of his experience alongside implementation; the work record distinguishes what he built from the programmes he led.
What can I inspect before getting in touch?: The MIT-licensed agent and skills library is public at https://github.com/vasilyu1983/AI-Agents-public. Cosmic Copilot and AEO Platform have live product sites. The library is a curated public slice, not the private services themselves. The work record explains the fraud detection engine and the governance retrieval capability; technical walkthroughs of private work are available on request, subject to confidentiality.

Dispatch · 22 Aug 2026

How I check AI-generated work

Task checks, independent review and human approval in my agent workflows.

Read the dispatch

Advisory · Alongside the day job

A defined problem to work through?

I also take a small number of scoped commissions in regulated fintech. The engagement shapes and boundaries are set out separately.

Read about advisory work

Get in touch

Let’s talk about what you’re building.

If your team needs hands-on AI work with product and business judgment, I’d welcome a conversation. A short note is a good place to start.

Get in touch