Skip to content
View Rafael215's full-sized avatar

Highlights

  • Pro

Block or report Rafael215

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 donโ€™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 userโ€™s behavior. Learn more about reporting abuse.

Report abuse
Rafael215/README.md

Hi, I'm Rafael ๐Ÿ‘‹

Iโ€™m a Computer Science student at UC San Diego with a minor in Cognitive Science, focused on software engineering, AI-driven products, and data-intensive applications. I enjoy building tools that solve practical problems, especially across healthcare, education, finance, and accessibility.


๐Ÿ”ง Tech Stack

Languages

  • Python, Java (Spring Boot), C/C++, TypeScript, JavaScript, SQL (Postgres), HTML/CSS

Frameworks & Libraries

  • Backend: FastAPI, Flask, Node.js, Next.js, Spring Boot
  • Frontend: React, React Native, Streamlit
  • Data/ML: pandas, NumPy, Matplotlib, Seaborn, scikit-learn
  • Testing: JUnit, PyTest

DevOps & Tools

  • AWS, Google Cloud Platform, Docker, Kubernetes, Linux/Unix, Supabase, Firebase
  • CI/CD pipelines, Git & GitHub for version control

๐Ÿ›  Selected Projects

๐Ÿพ Stranger Danger โ€” React, TypeScript, Supabase, Google Gemini, ElevenLabs

Multimodal AI application for real-time wildlife identification and outdoor learning.

  • Built a full-stack application that identifies wildlife from photo, audio, and video inputs using the Google Gemini 2.5 Flash API.
  • Developed interactive features including an AR field scanner, an AI survival simulator, and an audio training system powered by ElevenLabs.
  • Designed a geolocation-based community sightings feed and personal field guide backed by Supabase/Postgres for real-time updates and persistent species tracking.

Focus: multimodal AI, full-stack product design, and scalable edge-based architecture.


โšฝ Tactical Viewer App โ€” Python, SQL, Streamlit, Supabase, pandas, NumPy, Matplotlib

Interactive analytics platform for exploring soccer tracking data and tactical decision-making.

  • Engineered an ETL pipeline that downsampled 25 fps World Cup tracking data to 10 fps, filtered frames around key match events, and loaded millions of rows into a normalized Supabase/Postgres schema across 15+ tables.
  • Implemented Voronoi pitch control, grid-based expected threat (xT), and logistic regression pass probability models to quantify spatial dominance and passing danger per frame.
  • Built a Streamlit tactical viewer with real-time heatmaps, pass overlays, and frame-by-frame pitch control diagrams queryable by match and event.

Focus: sports analytics, data engineering, and interactive decision-support tooling.


โ˜• Compliance Autofill Engine (CAfE) โ€” Python, FastAPI, AWS Bedrock, React

AI-assisted system that converts advisor meeting notes into structured, audit-ready compliance documentation.

  • Built a full-stack AI-assisted workflow that transforms meeting notes and client profiles into structured compliance drafts, reducing manual write-up time from 20 minutes to 3 minutes.
  • Implemented document ingestion, retrieval over compliance guidance, and AWS Bedrock-powered structured JSON generation with post-processing for more consistent outputs.
  • Designed a human-in-the-loop review flow so advisors can inspect, edit, and approve each generated section before finalization.

Focus: applied NLP, backend system design, and human-centered AI.


๐Ÿ“ˆ Portfolio Analyzer โ€” Python, pandas, yfinance

Tool for analyzing personal investment portfolios using live market data.

  • Built a portfolio analysis system that computes asset allocation, unrealized profit/loss, benchmark returns, and risk-adjusted performance (Sharpe ratio).
  • Integrated live market data via yfinance, supporting configurable benchmarks, time horizons, and risk-free rates.
  • Designed both a CLI interface and an interactive dashboard, enabling flexible analysis across different investing strategies.

Focus: financial data analysis, probabilistic reasoning, and decision-support tooling.


๐Ÿ’ผ Experience & Leadership

VP External โ€” UCSD SHPE (Society of Hispanic Professional Engineers)

  • Secured $4,000+ in sponsorships from companies like LPL Financial, Turner, Bank of America, Lockheed Martin, and Northrop Grumman.
  • Organized high-impact workshops and networking events with 100+ attendees.
  • Designed logistics and mentorship coordination for Ideathon, improving mentorโ€“participant support.

UC San Diego Bookstore โ€” Receiving Associate

  • Verified, received, and routed incoming merchandise shipments while communicating quantity and quality discrepancies quickly and accurately.
  • Performed data entry and inventory updates in NetSuite and supported packing, labeling, and shipping to keep daily operations moving efficiently.

These roles strengthened my skills in communication, ownership, and cross-functional collaborationโ€”all crucial for building impactful software.


๐ŸŒŽ About Me

  • ๐ŸŽ“ CS @ UC San Diego, graduating 2028
  • ๐Ÿ‡ฒ๐Ÿ‡ฝ First-generation Mexican, U.S. citizen
  • ๐Ÿ—ฃ Bilingual: English, Spanish

Iโ€™m currently looking for Software Engineering Internship opportunities where I can contribute to real products, learn from experienced engineers, and keep growing as a developer.


๐Ÿ“ซ Letโ€™s Connect

If youโ€™re working on something at the intersection of software, data, finance, and human impact, Iโ€™d love to connect.

Pinned Loading

  1. portfolio-analyzer portfolio-analyzer Public

    CLI portfolio analytics tool that computes allocation, unrealized P/L, benchmark returns, and risk-adjusted performance using live market data.

    Python