Skip to content
Available for internships

Lukalapu Mohnish

Third year Mathematics and Computing at IISc Bengaluru. Retrieval systems, model compression, robotics simulation, and physics proved in Lean 4.

View projectsGitHub

I build retrieval systems that survive a rare token, compress vision transformers until they fit on a laptop, generate synthetic training data in Isaac Sim, and write physics in Lean 4 so the proofs ship with the code.

  • Retrieval and RAG
  • Model compression
  • Robotics simulation
  • Formal verification

Selected work

Three of seven. The note on each is the part that was actually hard, not the part that took longest.

Systems2026Project lead

Multi-Tenant Legal RAG Assistant

A legal-document retrieval system with two ingestion engines, hybrid search and per-session document isolation: a FastAPI backend behind a Cloudflare tunnel serving a static React SPA.

Retrieval fuses BM25 and dense-vector *ranks*, not scores. A cosine distance and a BM25 score share no scale, so any normalisation between them ends up tuned to one corpus and wrong on the next. The lexical leg exists because an embedding blurs exactly the rare token, “Section 302”, that made the question specific.

  • FastAPI
  • Celery
  • Redis
  • ChromaDB
  • BM25
  • +3
Read case studyMulti-Tenant Legal RAG Assistant source on GitHub
Kurt's tutor chat interface showing a grounded answer with an inline-rendered mathematical equation
AI/MLMay 2026 to Jul 2026Project lead

Kurt, an Offline-First AI Tutor

A fully local desktop assistant that transcribes lectures, answers questions grounded in your own documents, and compiles the result into typeset PDFs. No network required.

Tkinter has no MathJax, so multi-line equations are intercepted from the model's output and rendered through matplotlib's typesetting engine into transparent PNGs embedded inline: real mathematics in a desktop chat window, entirely offline.

  • Python
  • faster-whisper
  • Ollama
  • ChromaDB
  • LangChain
  • +3
Read case studyKurt, an Offline-First AI Tutor source on GitHub
WebJun 2026 to Jul 2026Project lead

F1 Telemetry Dashboard

Live and historical Formula 1 telemetry: speed, throttle, brake, RPM, gear, DRS, tyre strategy and a GPS track map, streaming from F1's official timing feed with no paid tier.

Drivers cross the line at different moments, so speed-against-time never lines up. Telemetry is interpolated onto a uniform five-metre distance grid, which is what makes two drivers genuinely comparable at the same braking point.

  • Streamlit
  • Plotly
  • FastF1
  • SignalR
  • pandas
  • +2
Read case studyF1 Telemetry Dashboard source on GitHub
All seven projects

What I work with

Four areas, following the same grouping the CV uses.

Machine Learning

Where most of the last two years went. Retrieval, transformers, and making models small enough to run on hardware I actually own.

PyTorch, Retrieval and RAG, Vision Transformers, Model Compression

Systems Engineering

The unglamorous half of a working project: services, queues, stores, isolation boundaries and tests.

Python, FastAPI, Git and Linux, Testing

Robotics and Simulation

Synthetic worlds and the physical ones they stand in for, from a rover chassis to an egocentric hand tracker.

SolidWorks, NVIDIA Isaac Sim, Isaac Lab, Simulation-to-Reality Alignment

Mathematics and Formal Methods

The degree, and the reason the compression work reads as mathematics rather than as trial and error.

Linear Algebra, Calculus, Probability and Statistics, Optimisation Theory

Every skill, with the evidence

Where I have worked

Two internships, both within the last year.

  1. May 2026 to Jul 2026

    emergence AI

    Research Intern

    Formal verification applied to a physics engine, so that reinforcement learning agents cannot farm rewards out of the simulator's own numerical mistakes.

  2. Dec 2025 to Mar 2026

    Saturn Labs

    Robotics Engineering Intern

    Synthetic data and simulation for egocentric vision: generating the training set in NVIDIA Isaac Sim, then making the model that consumed it fast enough to run live.

Positions, education and results

Looking for an internship in machine learning or robotics.

Happy to talk about any of the work on this site, and about research collaborations. Email reaches me fastest.

Get in touchmohnishl@iisc.ac.in