Harshavardhana Naganagoudar

▾ [About]

I'm an AI/ML engineer exploring reasoning models, autonomous agents, and generative systems. I enjoy building practical AI products across reinforcement learning, LLM workflows, and multimodal applications. After six years in software testing and automation, I transitioned into AI/ML to work on systems that combine intelligence, creativity, and real-world usability.

▾ [Projects]

Influo — AI-driven promotional video creation

Generates influencer-style short-form videos from product images using multimodal AI pipelines. Integrates React, FastAPI, and Ollama for end-to-end automated video generation. GitHub

Qwen2.5-3B GRPO — RL-based mathematical reasoning

Fine-tuned Qwen2.5-3B with LoRA + GRPO for arithmetic reasoning. Used custom reward shaping for structured equation generation, trained on Countdown arithmetic tasks. GitHub

Repo For Dummies — AI-powered repository onboarding

Two-agent architecture: a GitHub retrieval agent + an instruction-generation agent that simplifies onboarding for any open-source repository automatically. GitHub

▾ [Previously]

I previously worked at IBM as a Test Specialist, focusing on automation, backend testing, CI/CD pipelines, and release reliability for enterprise systems.

▾ [Interests]

I enjoy running, reading, anime, and puzzle games — anything that rewards curiosity and patience. A lot of my learning comes from exploring ideas online, especially listening to Andrej Karpathy somehow make neural networks feel surprisingly intuitive.