Hi, I'm Prabhaav :)

Computer Science Graduate Student & Software Engineer

I’m an M.S. in Computer Science student at Michigan State University specializing in full-stack web applications, real-time systems, and computer vision. From architecting enterprise React and .NET solutions used by thousands of professionals to deploying real-time C++ machine learning models on vehicles, I focus on building clean, production-grade software that solves real-world problems.

Projects

Vision ASM

Vision ASM

Open-source AI blind-spot monitoring for StarPilot vehicles.

PythonC++YOLOONNXComputer Vision
Vision ASMVision ASMVision ASM

A safety-critical, community-driven vision system that detects vehicles in adjacent lanes using the driver-facing wide-angle camera. It ships in StarPilot's stable release and runs in real time on embedded automotive hardware.

  • Appointed to Development Team with full repository write access to help steer project roadmap and backend architecture.
  • Authored and merged 1.7K+ lines of C++/Python into production, deploying a real-time blind-spot monitor to thousands of active vehicles.
  • Trained custom YOLO26n ONNX models on 27+ hours of crowdsourced driving video with semi-automated CAN-bus radar labeling.
  • Engineered edge performance on Snapdragon SoCs using zero-copy NV12 buffer slicing and CPU core isolation under 1.2x throttle overhead.
  • Drove community adoption (27K+ views, 175+ upvotes) and maintained the production lifecycle across bug fixes and dataset updates.
View Project
Movie Recommendation Engine

Movie Recommendation Engine

A from-scratch recommender system over IMDb data.

PythonWeb ScrapingEmbeddingsTypeScript
Movie Recommendation EngineMovie Recommendation Engine

Built for a CSE 881 data mining course project, this system recommends movies using data gathered entirely from IMDb. It emphasizes a full pipeline: acquisition, preprocessing, feature engineering, and similarity scoring.

  • Scraped and parsed raw IMDb datasets (titles, crew, ratings) plus custom HTML scraping for supplemental metadata.
  • Built text embeddings from plot summaries to power cosine-similarity recommendations.
  • Shipped an interactive frontend for browsing and comparing recommendations.
View Project
PortVoy

PortVoy

Android app to discover, scan, and connect to local servers.

JavaAndroidNetworking
PortVoy

A privacy-first Android utility that makes network scanning effortless. It inspects your network interface, scans a chosen port, and lets you connect: all with a simple interface and no data collection, tracking, or ads.

  • Discover network interfaces (including hotspot connections) and scan for open ports.
  • One-tap connect to frequently used servers like FrogPilot or Portainer via a settings shortcut.
  • Fully open source and on-device only - everything stays on your phone.
View Project
Lon-Capa Grade Calculator

Lon-Capa Grade Calculator

Grade calculator for Lon-Capa physics courses.

HTMLCSSJavaScript
Lon-Capa Grade Calculator

A web-based calculator that turns copy-pasted Lon-Capa data into a detailed grade breakdown. Customize weights per category and instantly see your overall percentage and GPA classification.

  • Paste lecture, homework, exam, and LearnSmart data in one go.
  • Adjust weightage per category with live recalculation.
  • Automatic grade & GPA classification with a modern, intuitive UI.
View Project
StudyLink

StudyLink

Campus social app to meet peers, find events, and join groups.

JavaAndroidFirebaseImgBB
StudyLink

An Android app built for a CSE 476 course project that helps students connect on campus: discover events and groups and find others in similar classes. Firebase-based account management keeps sign-up and data storage secure.

  • Connect with classmates, discover campus events, and join interest-based groups.
  • Firebase-powered authentication (create/delete accounts) and secure data storage.
  • Image saving/loading integrated via ImgBB; polished profile and homescreen flows.
View Project
Phantom Traffic Jam Study

Phantom Traffic Jam Study

Research on mitigating phantom traffic jams via CACC in mixed fleets.

JavaScriptPythonTraffic SimulationResearch
Phantom Traffic Jam StudyPhantom Traffic Jam Study

A collaborative study and multi-lane traffic simulation investigating whether Cooperative Adaptive Cruise Control (CACC) can act as a 'traffic shock absorber' in mixed fleets of human-driven and automated vehicles.

  • Modeled mixed-fleet traffic with IDM, CACC, and MOBIL in a ring-road simulation to reproduce phantom jams.
  • Varied CACC penetration from 0% to 100% across single- and double-lane scenarios, measuring flow stability and congestion.
  • Found meaningful stability gains from partial CACC adoption (peak improvement around 60%), informing V2X investment thresholds.
View Project

Papers

Adaptive LLM Load-Balanced Routing

Open

AV Penetration & Phantom Traffic Jams

Open

Learning vs. Heuristic Movie Recommendation

Open

Traffic Sign Physical Corruption

Open

Resume