Pragnyan Ramtha

about me.

Hello! My name is Pragnyan and I enjoy creating things that make people go Damn! My interest in programming started back in 2020 when I decided to try editing custom Linux distros, turns out hacking together a custom OS taught me a lot about Linux and Programming! Fast-forward to today, I have worked as a Founding Engineer at Six Axis Studios. I’m focused on building projects, contributing to open source, and working on my skills. I also recently joined NIAT - MRV University and would love to network with students and alumni.

experience.

• Founding Engineer Remote

  at, Six Axis Studios

  May, 2025 - Aug, 2025

  • Built an AI agent to handle customer support queries, reducing response times by 50% and improving customer satisfaction by 65%.
  • Found insights in client data to improve the company's sales by 20%.
  • Engineered internal AI agents for data analysis and marketing automation to increase efficiency and productivity of the company by 30%.

  • Scikit-learn
  • LLM Agents
  • APIs
  • Python

• Open Source Developer Remote

  at, Various Open Source Projects

  2024 - Present

  • Actively Contributed to 25+ open-source projects, across various organizations.
  • Refactored and enhanced the codebase to boost maintainability, achieving an 80% developer satisfaction rate.
  • Winner of IEEE Summer of Code (IEEESoC) Hackathon 2025, for my open source contribution to multiple projects.

  • Python
  • TypeScript
  • Git
  • Docker
  • CI/CD

projects.

• Personality Clone

  • Using contrastive fine-tuning methodology, I fine-tuned a Sentence Transformer (SBERT) on private conversational data to mimic my response style.
  • Implemented a siamese network architecture with cosine similarity loss, which improved semantic embeddings and achieved 92% accuracy in replicating my response style, a 28% improvement over baseline models.

  • TensorFlow
  • Python
  • CUDA
  • Transformers

• Autopilot

  • Using function calling and tool-use paradigms, I built an intelligent AI automation system that performs complex tasks using natural language commands and AI agents.
  • Implemented a Reasoning + Acting agent framework with safe command execution sandboxing, reducing execution errors and achieving 45% faster task completion compared to manual workflows.

  • Python
  • LLM Agents
  • APIs

• CheatSheet AI

  • Using retrieval-augmented generation (RAG) and chain-of-thought prompting, I developed an AI system that generates contextual cheat sheets for various topics.
  • Implemented semantic chunking with vector embeddings and multi-shot learning, improving content relevance and achieving 65% reduction in user study time compared to traditional learning methods.

  • Python
  • LLM Agents
  • APIs

technical blogs.

• Training a lightweight personality clone August 22, 2025

  Read more 6 min read

  Dataset prep from chats, contrastive fine-tuning, and evaluation pitfalls.

• How I’m building Autopilot September 5, 2025

  Read more 5 min read

  Notes on architecture choices, safe command execution, and early UX experiments.

• CheatSheet AI – ranking experiments September 1, 2025

  Read more 4 min read

  Trying hybrid retrieval + LLM reranking for better meta-search relevance.

technical skills.

• Languages:

  Python, TypeScript

• Certifications:

  Machine learning certification (Stanford), CS50: comp. Sci. (Harvard University)

• Libraries:

  Numpy, Pandas, Matplotlib, Scikit-learn, Pytorch, Transformers, Unsloth, Ollama

• Cloud & Deployment:

  Github, Vercel, self-hosting, Google Cloud Platform

• Achievements:

  • Winner, IEEE Summer of Code (IEEESOC) Hackathon 2025
  • Winner, Empathy Encryption Hackathon 2025
  • Winner, Daydream Hyderabad @ Hackclub 2025
  • Top 0.5% Finalist, Shell AI Hackathon 2025

• Tool kit:

  uv, archlinux, Git, Docker, Neovim