Constantine Samarikas

Los Angeles · University of Manchester, 2026

Constantine Samarikas

Data, AI and software, from financial-crime analytics to products people use.

Email LinkedIn GitHub

About

2026 graduate of the University of Manchester, with a First Class dissertation on detecting Bitcoin money laundering from regulatory red flags. Interested in applied machine learning and data that can be trusted.

Projects and research

Project · 2026 vendr.pro ↗

Vendr

Free collection, pricing and trading app for Pokémon and One Piece card collectors and shops in Europe.

  • 10card shops in Greece
  • 9European and US price sources, merged nightly
  • 3platforms: iOS, Android and web
  • Used by a few hundred collectors alongside the shops.
  • Every price change is checked against a second source before it is published.
  • Card identification and pricing from a photo.
  • TypeScript
  • React Native
  • PostgreSQL
  • GitHub Actions
Research · 2026 · First Class Code on GitHub ↗PDF ↗

Operationalising Regulatory Red Flags for Bitcoin Anti-Money Laundering Detection

BSc dissertation, The University of Manchester

Turns the six FATF virtual-asset red flags into 35 measurable features and detects illicit activity without large hand-labelled datasets, using rule-based pseudo-labelling, clustering and LightGBM on data from the Cardiff University Bitcoin Database.

  • 0.96macro F1 on a balanced evaluation set
  • 8.3Mtransactions, 24 quarterly snapshots, 2019–2024
  • 58.6% → 7.5%mixing-related flags, Q3 2023 to Q3 2024
  • 0.96 macro F1 on a balanced evaluation set of 225,722 transactions (the 149,128 confirmed labels plus a sample of unmatched transactions presumed licit); 0.93 on the confirmed labels alone.
  • Rule firing tracked enforcement: mixing-related flags peaked about a year after the Tornado Cash sanctions, then fell as Samourai and Wasabi faced enforcement.
  • Matched or exceeded published results on the Elliptic++ benchmark.

Background

2025

Deloitte, Athens

Digital forensics intern. Given the go-ahead to use AI in casework on condition of showing the team how: a redaction expected to take days took about five minutes with an AI-written Word macro run offline, so no case data left the machine. Selected for the firm's AI team.

2023–2026

The University of Manchester

BSc Information Technology Management for Business, Alliance Manchester Business School. First Class dissertation.

2024–2025

Client projects

Lloyds Banking Group (financial literacy app, sole developer), Procter & Gamble (AR customer-support prototype), Ford (UX research).

Tools · Python, SQL, TypeScript, React Native, PostgreSQL, scikit-learn, LightGBM, Claude Code

Contact

LinkedIn GitHub