• 🔭 I’m currently working on:

  • 🚀 Stable Diffusion Deployment using Flask, Docker, and RunPod with a custom API pipeline.
  • 🧠 LangGraph Agents with multi-agent memory workflows and OpenAI Assistants.
  • 🎮 Multithreaded OpenGL Physics Engine with GPU optimization and real-time simulation.
  • 🧩 Image Segmentation with Mask2Former on custom datasets for fine-grained instance masks.
  • 🤖 Finetuning Qwen-1.5B using GRPO (Generalized Reinforcement Pretraining Objective) for task adaptation.

• 🌱 I’m learning advanced optimization, multithreading in C++, RLHF, and MLOps best practices.

• 💬 Ask me about:

  • ML/AI deployment with Flask, Gunicorn, and PyTorch
  • Full-stack apps using Django + React + PostgreSQL
  • Robotics, real-time sensors, and cloud-based pipelines

• 📫 Reach me at: bhupathinikhil2006@gmail.com

• 🧠 Stable Diffusion API (GPU Inference)
  Cloud-based model deployment via RunPod with Flask + Gunicorn
  → inference | logging | security | Dockerized endpoint

• 🧩 LangGraph Agents
  Multi-agent system for automating web tasks using OpenAI functions + Firecrawl
  → memory streaming | UVICORN backend | Markdown + logs

• 🧲 OpenGL Physics Engine
  Parallel C++ simulation with sphere collisions, GPU threads & constraints
  → real-time rendering | friction | UI controls

• 🔬 Image Segmentation with Mask2Former
  Training instance-aware segmentation models on custom datasets
  → Detectron2 | Mask2Former | data pipelines | metrics

• 🧠 Finetuning Qwen-1.5B using GRPO
  Experimenting with reinforcement pretraining on dialogue and instruction tasks
  → LLM tuning | Hugging Face | GRPO | RL-style reward shaping

🧠 “Code is poetry when it solves a real-world problem efficiently.”