Shlok Mehndiratta

Projects & Coursework

A collection of projects I have built and courses I have taken during my degree at IISER Bhopal. I enjoy working at the intersection of robotics, control, computer vision, and software, taking mathematical ideas and turning them into working systems.

Featured Projects

Systems I have designed and built across robotics, vision, and software.

01 Robotics & Control Systems
2025 – 2026

Robust Linear Control: A Belief-Coupled Tube-MPC Framework

Linear Control Systems (ECS 334) • IISER Bhopal

Developed a unified research framework for autonomous tracking that bridges robust control theory, high-dynamic flight trajectory validation, and closed-loop ROS 2 / Gazebo simulation.

ROS 2 Jazzy Gazebo Harmonic Tube-MPC Robust Control (RPI Sets) EKF Crazyflie SITL Docker C++ Python
View details & architecture
What I built & tested
  • • Mathematical Foundations: Derived tracking feedback gains and tightened nominal state/input constraints using Robust Positively Invariant (RPI) sets to guarantee stability under bounded external disturbances.
  • • Trajectory Engineering: Validated high-dynamic 3D flight trajectories for micro-UAV hardware using Crazyflie SITL simulation, containerized with Docker for full reproducibility.
  • • System Integration & Closed-Loop Control: Integrated perception (OpenCV), state estimation (Extended Kalman Filter), and belief-space control within modular ROS 2 nodes in Gazebo Harmonic.
  • • Adaptive Tracking: Evaluated Adaptive Tube-MPC and belief-coupled tracking against standard baseline controllers under sensor noise and dynamic curvature.
Key takeaway: Guaranteed trajectory tracking stability within bounded disturbance tubes while closing the loop between perception, state estimation, and predictive control.
02 Embedded Systems & Hardware
Oct – Nov 2025

Digital Logic Design & Bare-Metal AVR Microcontroller

IIT Hyderabad (with Dr. G V V Sharma) / IISER Bhopal

Programmed an ATmega328P microcontroller using bare-metal AVR assembly and direct hardware register manipulation without external runtime libraries, alongside discrete digital logic prototyping.

AVR Assembly Direct Register I/O ATmega328P Digital Logic Circuit Design
View details & architecture
What I built & tested
  • • Wrote bare-metal AVR assembly from scratch, directly writing to hardware control registers to manage clock timers, pin interrupts, and I/O ports.
  • • Designed combinational and synchronous digital building blocks including decade counters, 4-to-7 line decoders, and discrete logic gates.
  • • Wired physical breadboard circuits to drive multiplexed seven-segment displays with cycle-accurate hardware timing.
Key takeaway: Built a concrete, machine-level foundation in microcontroller execution, memory-mapped I/O, and hardware timing constraints.
03 Computer Vision & ML
Dec 2024 – Jan 2025

3D Point Cloud Segmentation with Graph Neural Networks

GeoAI4Cities Lab (Advisor: Dr. Vaibhav Kumar)

I developed a machine learning pipeline that uses graph neural networks to label large 3D urban LiDAR scans, helping computers tell buildings, roads, and trees apart.

PyTorch Open3D RANSAC Python 3D Vision
View details & architecture
What I built & tested
  • • Represented 3D point clouds as graphs where each point connects to its nearest geometric neighbors to capture surrounding shapes.
  • • Applied RANSAC geometric filtering to clean up sensor reflections and stray noise before running predictions.
  • • Trained a 3D graph neural network in PyTorch, outperforming standard point-by-point classifiers in complex city scenes.
Key takeaway: Showed that learning from local geometric neighbors produces much cleaner 3D boundaries than treating points in isolation.
04 Software Systems & Web Architecture
Jan 2024 – Jun 2025

Full-Stack Web Systems Suite (Harvard CS50 Web)

Harvard University CS50 Web

Engineered a portfolio of five full-stack web applications spanning frontend semantics, relational database modeling, asynchronous single-page applications, and session-based authentication.

Django Python 3 JavaScript (ES6) SQLite / PostgreSQL REST APIs HTML5/CSS3
View details & architecture
What I built & tested
  • • Single-Page Email Client: Built an asynchronous SPA in vanilla JavaScript that manages inbox views, dynamic mail archives, and inline replies via RESTful APIs without page reloads.
  • • E-Commerce Auction Platform: Developed an online bidding marketplace in Django and SQLite with user watchlists, category filtering, and real-time bid validation.
  • • Wiki Encyclopedia & Social Network: Implemented an editable Markdown encyclopedia engine and a paginated social network with follower graph feeds.
  • • Foundational Frontend: Replicated Google Search interfaces covering HTML5 semantics, CSS3 flexbox layouts, and accessibility standards.
Key takeaway: Mastered core backend architectural patterns, relational database constraints, and RESTful API state management from first principles.
05 AI & Web Applications
2024

AI-Powered Car Lease & Loan Contract Assistant

Infosys Virtual Internship Program 6.0

Engineered a mobile-friendly AI application that extracts key financial terms from auto lease and loan contracts, calculates a deal fairness score, and provides turn-by-turn negotiation strategies.

FastAPI Python 3.12 Ollama (Llama 3.2) LangChain Flutter / Dart Docker NHTSA APIs Tesseract OCR
View details & architecture
What I built & tested
  • • Dual-Pipeline Contract Analysis: Combined rule-based regex and local LLM extraction (Ollama Llama 3.2 via LangChain) to parse APR, down payment, residual values, and hidden fee clauses from PDF contracts.
  • • Deal Fairness Engine: Implemented a proprietary fairness scoring model (0–100) and vehicle depreciation estimator built on a 10-year MSRP database with age/mileage curves.
  • • NHTSA Vehicle Verification: Connected real-time NHTSA APIs to decode vehicle VIN numbers, cross-referencing manufacturer specifications against safety recall and complaint records.
  • • Interactive Negotiation Chatbot: Built an LLM-driven negotiation assistant that analyzes contract leverage points to guide users and draft professional dealer counter-offer emails.
  • • Cross-Platform Stack: Built a containerized FastAPI backend deployed with Docker, paired with a clean, mobile-first Flutter interface.
Key takeaway: Eliminated dealer-consumer information asymmetry by turning complex auto finance agreements into clear, structured terms and negotiation leverage.

Academic Coursework

47 Courses Total

Current ongoing courses and completed curriculum tracks from my Electrical Engineering & Computer Science degree at IISER Bhopal.

IISER Bhopal EECS
Curriculum Tracks

Robotics, Autonomy & Control Systems

Feedback control, state-space models, system identification, and physical robot experiments.

4 Courses

Control Systems

ECS323

Classical feedback loops, stability margins, Bode and Nyquist criteria, and PID controller tuning.

Linear Control Systems

ECS334

State-space formulation, controllability and observability matrices, and state feedback controller synthesis.

Signals and Systems

ECS204

Continuous and discrete-time signals, Fourier and Laplace transforms, frequency response, and filtering.

Principles of Communications

ECS301

Analog and digital modulation schemes, signal bandwidth, noise analysis, and data transmission protocols.

Certifications & Training

Practical workshops and certified programs I have completed alongside my degree.

3 Programs
C-DAC & IIT Roorkee (Supported by MeitY) June 2025

Quantum Computing Program

CDACH/QML/275

Completed a 4-week program on Quantum Computing, jointly conducted by C-DAC and IIT Roorkee with support from MeitY. Covered fundamentals of qubits, quantum gates, entanglement, and teleportation, along with hands-on practice in quantum simulators, quantum algorithms (Deutsch–Jozsa, Simon’s, Grover’s, HHL), and an introduction to Quantum Machine Learning (QML).

Quantum Computing Qubit Gates & Circuits Grover & HHL Algorithms Quantum Simulators QML Introduction Python
Credential CDACH/QML/275
Harvard University (edX) 2024 – 2025

CS50's Web Programming with Python and JavaScript

A hands-on software engineering program from Harvard. Covered building complete web applications from scratch, including Django backends, relational SQL databases, interactive JavaScript frontends, user authentication, automated testing, and web security.

Django Python JavaScript SQL & Django ORM REST APIs CI/CD & Testing Web Security
Read my notebook reflection →
Infosys Springboard 2024

Infosys Springboard Virtual Internship 6.0

Virtual Internship 6.0

Completed a virtual internship focused on developing practical AI applications. Built the Car Lease and Loan Contract Review Assistant, combining large language models for clause extraction with a FastAPI backend and a Flutter mobile app.

Large Language Models (LLMs) FastAPI Flutter Docker Python REST APIs
Verified Track Completion