Projects

A full-stack research project built during my BII Grant @ FEUP. An automated pipeline that scrapes, parses, and semantically audits university Course Information Sheets using a multi-provider AI factory.

• Designed a robust scraping pipeline handling malformed and inconsistent HTML
• Built a provider-agnostic AI factory (OpenAI, Anthropic, local models) for flexible LLM calls

Real-time postural feedback for exercises using Computer Vision. Inspired by volleyball — helps athletes improve technique in a private, non-judgmental environment.

• Understood MediaPipe's pose estimation model and how to map 33 skeleton keypoints to angles
• Built real-time angle calculation for squat depth, plank alignment, and lunge form

An AI-powered tool that customises your resume to match a target job description — intelligently rewriting and repositioning your experience to maximise relevance.

• Worked with PDF parsing and text extraction in server-side TypeScript
• Designed prompts that rewrite faithfully — preserving voice while improving relevance

A retro-style game built under the strict technical constraints of PICO-8 — 128×128 display, 128 sprites, 32KB cartridge. Won 1st place at RetroJam.

• Built a complete game loop in Lua: input, physics, state machine, collision, rendering
• Learned that extreme constraints force creativity in ways total freedom never does

Low-level K-Means clustering in Assembly with real-time hardware visualisation on a simulated 32×32 LED matrix. Focused on register-level efficiency and memory management.

• Understood exactly how high-level algorithms translate to register operations
• Managed memory manually with no safety net — pointer arithmetic by hand

A pipeline for aggregating and processing feedback at scale, helping teams surface the signals that matter most from large volumes of unstructured input.

• Thought carefully about data transformation and pipeline composability
• Applied TypeScript generics for flexible, type-safe data processing stages