Skip to content
View pieroevcc's full-sized avatar
🎯
Focusing
🎯
Focusing
  • Virginia, USA
  • 04:50 (UTC -12:00)

Highlights

  • Pro

Block or report pieroevcc

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
pieroevcc/README.md

Piero Evangelista

Computer Engineering @ University of Virginia · Class of 2029


👋 Who I am

I'm a Computer Engineering student at UVA who learns by building. I care about software that solves a concrete problem and measurably improves on what it replaces: faster workflows, clearer data, less friction for the user.

🛠️ What I've worked on

  • Raytracer - CUDA ray tracer ported from RTIOW CPU reference, achieving 115x speedup on RTX 4060 by leveraging GPU parallelism for high-performance rendering.
  • ProbabilityRAG - Local-first RAG tutor over Grinstead & Snell probability text that combines hybrid semantic search with intelligent reranking to provide accurate, contextual answers without external API calls.
  • Basketball Shot Tracker - Real-time multiplayer web app where kids play a tabletop basketball game and turn their shots into heat maps and stats, teaching data science concepts through play.
  • MNISTDigitsToGIF - Video diffusion model for generating MNIST digit animations on a consumer GPU, implemented from scratch with PyTorch to explore efficient generative modeling.
  • SpeakToDeck - Locally transcribe and translate your audio, then import it straight into your Anki deck for seamless language learning without cloud dependencies.
  • LensTranslate - Paste a screenshot or TikTok link to get instant translations overlaid on the image, using OCR and real-time translation to convert foreign text in place across 6 languages.
  • Benefits Digger (hackathon) - Full-stack benefits-eligibility screener with explainable AI that breaks down why users qualify for specific benefits using structured logic and LLM reasoning.

🧰 Tools I use

  • Languages: Java · C/C++ · Python · TypeScript · SQL · HTML/CSS
  • AI/ML: RAG · PyTorch · Diffusion Models · Qdrant · BGE-M3 · Whisper · faster-whisper · LLM Guardrails
  • Frameworks: React.js · Node.js · FastAPI · React Native (Expo) · Streamlit · REST APIs
  • Developer Tools: Git · GitHub Actions · Docker · Firebase · SQLite · AWS · Playwright · pytest · Vite
  • AI Developer Tools: Claude Code · MCPs · Skills · Agents

📫 How to reach me

Pinned Loading

  1. Raytracer Raytracer Public

    CUDA ray tracer ported from RTIOW CPU reference, benchmarked on RTX 4060 with 115x speedup.

    Cuda

  2. ProbabilityRAG ProbabilityRAG Public

    Local-first RAG tutor over Grinstead & Snell probability text (BGE-M3 hybrid + reranker, Qdrant, Ollama, FastAPI/React).

    Python

  3. Basketball-Shot-Tracker Basketball-Shot-Tracker Public

    Real-time multiplayer web app where kids play a tabletop basketball game and turn their shots into heat maps and stats — teaching data science by playing. React + TypeScript + Firebase.

    TypeScript

  4. MNISTDigitsToGIF MNISTDigitsToGIF Public

    Video diffusion model for generating MNIST digit animations on a consumer GPU. Implemented from scratch with PyTorch.

    Python

  5. LensTranslate LensTranslate Public

    Paste a screenshot or TikTok link to get instant translations overlaid on the image. LensTranslate uses OCR and real-time translation to convert foreign text in place, supporting 6 languages across…

    TypeScript

  6. SpeakToDeck SpeakToDeck Public

    Locally transcribe and translate your audio, then import it straight into your Anki deck.

    Python