Madhav

Hello, I'm Madhav Kataria.

Advancing artificial intelligence and building cool software.

About Me

Profile Photo

Hi! I'm am Madhav Kataria, a passionate software engineer and AI enthusiast. Pre-Final year undergraduate at IIT Jodhpur, with a strong interest in machine learning, MLOps, and building impactful tech products. I love working at the intersection of software engineering and AI, where I can turn cutting-edge research into practical, scalable solutions.

When I'm not coding, you can find me exploring new technologies, contributing to open-source projects, or enjoying a good book. I'm always eager to connect with like-minded individuals and collaborate on exciting projects.

Contact View CV

What am I up to?

AI Research

Exploring the frontiers of artificial intelligence, focusing on perception and intelligent systems as an undergraduate researcher.

Building Software

I enjoy building software products that solve everyday problems. Check out some of my favorites!

Books

Reads some books

Software I’ve Built

Here are some of the projects I'm proud to have worked on. Explore them to see my skills in action.

AI Image Explainability
AI Image Explainability
VLMs
GANs
KANs
VAEs
LoRA
Knowledge Distillation
Diffusion Models
Built an AI that spots fake images and explains why — even in blurry 32×32 pixels. Combined smart CNNs with a vision-language model to detect artifacts and reveal the truth behind AI-generated visuals.
GitHub
DeepPlay: Autonomous Football Agents
DeepPlay: Autonomous Football Agents
Unity Engine
DQN
PPO
Blender
PyTorch
Reinforcement Learning
Leveraged Deep Q-Networks in PyTorch and TensorFlow to train Unity-based football agents from scratch, achieving a 40% boost in strategic decision-making accuracy and an 85% success rate in mastering complex gameplay strategies. A 3D Unity simulation visualizes learned agent behaviors and tactics.
GitHub Live Site
AI Agent for Domain-Specific QA
AI Agent for Domain-Specific QA
Prompt engineering
CoTs
Multi-task learning
Model distillation
Built a high-efficiency AI agent for domain-specific question answering using GPT-4, combining multi-task learning with prompt engineering and CoT reasoning to streamline support workflows, cut response times, and deliver scalable cost savings.
GitHub Live Site
DeepFusion-C
DeepFusion-C
C
CNN
MNIST
Model Adapter
Custom IDX-format MNIST loader
DeepFusion-C is a hand-optimized C implementation of a four-layer CNN for MNIST digit recognition, leveraging SIMD intrinsics and OpenMP to deliver sub-millisecond, >99.2%-accurate inference with detailed latency, throughput and memory profiling.
GitHub
View All Projects

Research Interests

Delving into the frontiers of technology to innovate and discover. Here are some of my primary areas of research focus.

Artificial Intelligence & Machine Learning
Exploring advanced algorithms and their applications in solving real-world problems, with a focus on ethical AI development.
Human-Computer Interaction
Investigating how users interact with technology to design more intuitive, effective, and enjoyable digital experiences.
Distributed Systems & Scalability
Researching robust and scalable architectures for distributed applications, ensuring high availability and performance.