RobbieWalmsley

Backend and ML pipeline for a real-time dance analysis product at a stealth-stage UK startup. A causal Transformer on 3D pose data powers movement scoring and live coaching feedback, deployed through MediaPipe, WebSockets, and AWS Lambda for sub-frame latency.

My contributionDesigned and built the ML pipeline and backend services — from the causal Transformer architecture and pose representation through the serverless inference path and WebSocket delivery layer.

A physical trash can with a built-in camera that classifies and sorts waste in real time using a DINO-pretrained ViT ensembled with a Swin Transformer, trained on the RealWaste dataset across nine waste categories. Achieved 95.5% multi-class and 98.7% binary recyclability accuracy. Built for a high-school NeurIPS competition.

My contributionBuilt the entire AI and software stack — the ViT + Swin Transformer ensemble, training pipeline, real-time classification, and control software. Teammates handled the physical Raspberry Pi build, camera rig, servo, and ultrasonic sensor.

A standardised plug-and-play framework for the Minutemen Alternative Investment Fund at UMass. Provides a unified pipeline for testing trading strategies across historical data, Monte Carlo simulations, and multiple data providers, catching overfitting through GBM and GAN-generated stress tests. Dual backtesting engines (bar-based and event-driven), Optuna parameter optimisation, and standardised HTML scorecards.

My contributionShipped the initial framework — the bar backtester, four-page scorecard generator, synthetic-data stress-test module, Optuna optimisation layer, and all ten data-provider integrations (Yahoo, Alpaca, Polygon, Tiingo, Finnhub, Alpha Vantage, Twelve Data, MarketStack, Stooq, FMP).

An AI-native 3D creation engine that lets anyone go from idea to playable experience in minutes. Users describe what they want — a mechanic, a level, character behaviour — and Razorbill generates a working project: scenes, scripts, prefabs, and interactive logic.

My contributionBuilt the 3D asset generator, terrain and world generation, animation system, and particle system. Optimised physics using Jolt. Designed the combat system and enemy AI with navmesh pathfinding. Improved the LLM layer with context caching, routing, and tool calling.

A solo-built iOS, iPadOS, and visionOS app combining fitness tracking and personal finance in a single tool. An on-device LLM integrates across health, spending, and card data to surface personalised insights — no data ever leaves the device. Free, no subscriptions, no ads. Available on the App Store.