Benedict Anokye-Davies

Benedict Anokye-Davies

CS & AI Student

Building projects at the intersection of software engineering, quantitative finance, and AI.

About

I'm a first-year Computer Science and AI student at the University of Nottingham, Bloomberg x Sutton Trust bursary recipient, and Sutton Trust STEM Scholar. I build production systems that go beyond coursework: a governed AI payments agent, a commissioned event-market trading platform, open-source infrastructure tooling, and evolutionary simulations.

Currently seeking software engineering and AI/ML engineering internships where I can contribute to production systems and learn from experienced engineers.

Experience

  • 2026

    Bloomberg x Sutton Trust Bursary
    Bloomberg

    Selected for a competitive engineering-focused programme. Insight week at Bloomberg London, July 2026.

  • 2025 — Present

    Computer Science with AI
    University of Nottingham

    First-class result in Assembly Language Programming. Studying algorithms, data structures, databases, AI fundamentals, and mathematics for computer science. Building trading systems and simulations as personal projects alongside coursework.

  • 2023 — Present

    STEM Scholar
    Sutton Trust

    Selected for national programme supporting high-potential students from low-income backgrounds into STEM careers. Ongoing mentoring, career development, and alumni network.

  • 2023

    IT Department
    SUEZ

    Assisted the SQL data team with operational tasks including query writing and database maintenance. Gained practical exposure to enterprise data workflows and team-based development processes.

Projects

  • Arbiter: governed payments agent

    An AI payments agent designed to make unapproved spending impossible. Every payment clears 12 deterministic first-match rules before any LLM is consulted, and a single governed settle() call is the only route to execution. Decisions are written to an append-only audit ledger. I own the backend and governance core; a partner built the Stripe webhook and dashboard. Built for the Nous x NVIDIA x Stripe hackathon.

    • Python
    • FastAPI
    • MCP
    • Stripe
    • pytest

  • helios-vps-runner

    Open-source CLI for safe live operations on remote servers: host allowlist, forbidden-content scanning, a two-stage apply gate, and an append-only audit trail. MIT-licensed, with a 28-test offline suite and CI green on Python 3.10 to 3.12.

    • Python
    • GitHub Actions
    • CI/CD
    • MIT

  • Event-Market Trading Platform

    Commissioned backend for an event-market trading product. Automated trading on NBA and NCAAB markets on Kalshi, a CFTC-regulated exchange, with real capital. Built the live order execution path, position reconciliation, stop-loss and circuit-breaker risk controls, and a FastAPI operator dashboard. Deployed on a Linux VPS under systemd with a behavioural regression suite.

    • Python
    • FastAPI
    • pytest
    • PostgreSQL
    • Linux
    • systemd
    • Nginx

  • Quantihack: algorithmic trading strategies

    Competition entry for the Quantihack algorithmic trading challenge. Five strategy slots cover trend following, inventory-aware market making, momentum breakouts, regime adaptation, and volatility spikes, with ATR risk controls and position limits.

    • Python
    • Regime detection
    • Risk controls

  • Moodle Sync

    Personal automation tool that mirrors my University of Nottingham Moodle modules to local storage. Authenticates via the Moodle mobile token endpoint, walks every enrolled course, downloads slides and resources, and tracks coursework deadlines. Runs nightly via cron with incremental sync state so only new content is fetched.

    • Python
    • requests
    • Moodle Web Services API
    • cron

Skills

Languages

  • Python
  • TypeScript
  • Java
  • C
  • ARM Assembly
  • SQL
  • HTML/CSS

Tools & Frameworks

  • asyncio
  • pytest
  • GitHub Actions
  • Anthropic API
  • OpenAI API

Infrastructure

  • Linux
  • systemd
  • Docker
  • Git
  • CI/CD
  • SSH

Data

  • SQLite
  • PostgreSQL