Skip to content
View Drabblesaur's full-sized avatar
😴
Programming while sleepy
😴
Programming while sleepy
  • San Jose, CA

Highlights

  • Pro

Organizations

@DevelopersGuild

Block or report Drabblesaur

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
Drabblesaur/README.md

Hi, I'm Johnny To πŸ‘‹

Software Engineer & Applied AI Enthusiast | MS Software Engineering @ SJSU


πŸš€ About Me

  • πŸŽ“ MS in Software Engineering (Data Science specialization) at San Jose State University, graduating May 2026
  • 🐻 BS in Computer Science from UC Riverside
  • πŸ€– Building LLM systems, RAG pipelines, and full-stack apps
  • πŸ”­ Currently exploring agentic tool orchestration and multi-source retrieval systems
  • 🌱 Interested in applied AI, distributed systems, and mobile/web development
  • πŸ’¬ Ask me about LangChain, RAG architectures, or full-stack microservices

πŸ› οΈ Tech Stack

Languages Python JavaScript SQL

Frameworks React React Native Next.js Express.js

AI / ML LangChain OpenAI FAISS Ollama

Data Science NumPy Pandas Jupyter

Cloud / DevOps AWS Docker

Databases PostgreSQL MongoDB


πŸ“Œ Featured Projects

Python LangChain FAISS OpenAI API

  • Architected an LLM-based agent system with tool orchestration, multi-source RAG, and query planning to answer SJSU course planning and degree audit questions
  • Built a dynamic tool-routing pipeline for selecting between database queries, retrieval systems, and constrained web search
  • Benchmarked Qwen 1.7B vs 14B models, surfacing latency/quality tradeoffs (20–30s vs 60–100s inference)
  • Designed secure tool-scoped data access to prevent leakage of sensitive student information under adversarial prompts

Python Node.js React Native PostgreSQL FastAPI

  • Architected a full-stack microservice carpooling platform for SJSU students spanning 11 independent services (matching, routing, payments, real-time safety monitoring)
  • Implemented a three-stage rider-driver matching pipeline (PostGIS proximity filtering + heterogeneous network embeddings + weighted cost function), achieving 84%+ match accuracy across 100-driver candidate pools
  • Engineered real-time safety monitoring that detects route deviations and speed anomalies via live GPS vs. planned route polylines, triggering automated alerts

πŸ“Š GitHub Stats

Johnny's GitHub stats

Johnny's GitHub streak


πŸ“« Reach me at tojohnny1559@gmail.com

Pinned Loading

  1. SJSU_VAssistant SJSU_VAssistant Public

    San Jose State CMPE 259 Final Project. University Course Advisor & Degree Audit Virtual Assistant

    Jupyter Notebook

  2. manamsriram/SJSU_Ridesharing manamsriram/SJSU_Ridesharing Public

    Cheaper and more eco-friendly way to decrease the number of more cars on the road and solve SJSU parking issues.

    Swift

  3. OfflineRAGNotes OfflineRAGNotes Public

    TypeScript

  4. JTSite JTSite Public

    JavaScript