AI/ML Engineer • Applied Researcher • Data Scientist

I build intelligent systems that run in the real world — not just in a Jupyter notebook.

• 🤖 PoseCorrect: Real-time rehab assistant that watches human movement, understands it, and talks back using LLMs + vision + biomechanical knowledge.
• 💹 Spectra: Quantum+Transformer forecasting pipeline for stock index movement and options strategy planning.
• ☁️ FluidCloud: Decentralized personal cloud that splits, encrypts, and distributes your data across your own devices.

I work across:

• Multimodal AI (vision + motion + language)
• LLMs with retrieval (RAG, MCP, agentic pipelines)
• Time series forecasting (classical, deep, and quantum)
• Distributed / edge systems

• Building realtime movement understanding for personalized rehab coaching (PoseCorrect / LymphFit)
• Pushing low-latency financial prediction models for high-frequency use (Spectra)
• Designing privacy-first personal cloud infrastructure with no centralized server (FluidCloud)

AI / ML: PyTorch, TensorFlow, Transformers, MediaPipe, OpenCV, Qiskit
Intelligence layer: RAG, LangChain, FAISS, Knowledge Graphs, MCP
Backend / Infra: Python, Rust, FastAPI, Flask, AWS (Lambda / Textract / S3), Docker
Other: Time-series modeling, EMG signal analysis, realtime streaming

• Email: abhihk02@gmail.com
• LinkedIn: linkedin.com/in/abhinavkochar
• My CV: Google Drive Link

• 🥈 2nd Place – Quantum Computing Track, UMKC Researchathon (Spectra)
• 🎖 Honorable Mention – Doctoral AI Track, UMKC Researchathon (PoseCorrect)
• 🏅 Exemplary Performance Award – Cloud Computing (CloudExpense)
• 🎓 Graduate Research Assistant, NIH-funded clinical AI project

I'm actively looking for research / applied ML roles (vision, LLMs, intelligent systems).
If you're doing something interesting in AI + real-world signal data, let’s talk.