Hello, I'm
Raghav

Computer Engineering @ University of Toronto

building cool stuff that works

View My Work Get In Touch

About Me

I’m a Computer Engineering student obsessed with building smart software. Most of my time goes into large language models, embeddings, and figuring out how to make inference faster.

I’m here to build high-impact AI systems. My goal is to push large models to their limits and make them faster, more efficient, and ready for real-world scale.

Technical Skills

Programming Languages

Python JavaScript C++ Java C SQL Rust Go

Web Technologies

React Node.js HTML/CSS Express.js Django Flask GraphQL MongoDB PostgreSQL Spring Boot FastAPI TypeScript

Tools & Technologies

Git Docker AWS Linux Docker Kubernetes CI/CD Jenkins Terraform

Machine Learning

TensorFlow PyTorch Scikit-learn OpenCV NLP OpenVINO Hugging Face Keras Pandas

Experience

Google Summer of Code Intel (OpenVINO Toolkit)

Machine Learning Open-Source Contributor May 2025 - Aug 2025
  • Developed Multimodal Embeddings system based on Openvino Model Server
  • Converting models from Hugging Face and ONNX to OpenVINO format
  • Optimized inference performance for large language models
  • Integrated Vector Database with gRPC and REST scripts to enable semantic search
  • Participated in code reviews and agile development processes

Datacurve

Y-Combinator

Software Engineering Intern June 2025 - August 2025
  • Engineered backend microservices for enterprise AI agents, processing 200k+ fragmented data entries/day and generating actionable insights, improving decision-making speed by 30%.
  • Contributed 5+ production-ready PRs to high-profile open-source repositories advancing LLM coding capabilities, increasing AI agent code coverage by 20%
  • Optimized data processing workflows, cutting average inference latency from 450ms → 60ms (7.5× speedup) and integrating Google and Okta APIs to reduce setup time by 25%.

University of Toronto Entrepreneurship Hatchery

Software Engineering Intern May 2024 - August 2024
  • Supporting multiple early-stage startups at the Hatchery accelerator
  • Built a RAG pipeline over 100K+ documents using LangChain, Pinecone, and FastAPI
  • Reduced API latency from 1100ms to 180ms using streaming and async execution
  • Fine-tuned LLaMA 2 with LoRA for document classification, improving task accuracy by 30%
  • Deployed scalable ML services with Docker and CI/CD, serving 20K+ monthly queries

Projects

ShellPilot

AI tool for devs to use Linux with ease

Python LLMS OpenVINO Flask
Code Live

RebuilderAI

AI powered resume analyzer

Python Gemini Sqlite3 JavaScript
Code Live

Multimodal Embeddings

An interactive dashboard for visualizing complex datasets with real-time updates. Features include custom charts, filtering capabilities, and export functionality.

React D3.js Node.js MongoDB
Code Live

GeoGuessr

An online game which lets user upload locations and guess them and get rankings based on them

Arduino Raspberry Pi Python MQTT
Code Demo

Publications

Transformer Architecture from scratch

Implemented the full Transformer model step by step in PyTorch, explaining each component (attention, embeddings, feedforward, encoder/decoder). Designed to demystify the architecture powering GPT, BERT, and Stable Diffusion.

Medium

Multimodal Embeddings based on OVMS

A blog authored by me and published by Intel OpenVINO Toolkit, showcasing my GSoC 2025 project. It explains the architecture and implementation of a high-throughput multimodal embeddings system integrating CLIP, DINO, and LAION models with OpenVINO Model Server, enabling real-time image-to-image search with a 30× inference speedup.

Medium

Resume

Raghav Yadav

Computer Engineering Student & Developer

Computer Engineering Student
Full-Stack Developer
Machine Learning Enthusiast
System Architecture
3+ Years Experience
15+ Projects Completed
40+ Technologies

Download My Resume

Get a detailed view of my experience, skills, and achievements in a professional format.

Download PDF View Online
Last Updated: August 2025
Format: PDF (1 Page)
Language: English

Get In Touch

Let's Connect

I'm always interested in new opportunities, collaborations, and interesting projects. Feel free to reach out!

raghav.yadav@mail.utoronto.ca
+1 (647) 892-4001
Toronto, ON