I build practical AI systems that combine reasoning, retrieval, orchestration, and deployment. My focus is on Agentic AI, Generative AI, LLM fine-tuning, and AI engineering — with a bias toward shipping useful products instead of flashy demos.
Multi-agent workflows, tool use, planning, orchestration, autonomous reasoning 🔎 RAG Systems
Retrieval pipelines, semantic search, reranking, document assistants, knowledge grounding
LoRA, QLoRA, instruction tuning, compact model adaptation, domain specialization ⚙️ AI Engineering
FastAPI, Streamlit, Docker, Hugging Face, deployment workflows, inference basics
An AI-powered investment research app built with a multi-agent LangGraph workflow to generate structured financial insights.
A compact model fine-tuning project for medical Q&A using QLoRA and efficient domain adaptation techniques.
A real-time automatic license plate recognition system combining detection and OCR for practical CV workflows.
A creative AI project focused on building a distinctive pixel-art style persona experience with a modern visual identity.
📚 storyGPT
A custom decoder-only Transformer project for story generation, showing hands-on work with model training and LLM fundamentals.
More experiments, AI prototypes, and learning projects are available across my repositories.
▸ Advanced multi-agent orchestration with LangGraph
▸ Better RAG architecture and retrieval quality
▸ LLM fine-tuning and instruction-following behavior
▸ Inference optimization and deployment workflows
▸ Multimodal AI and next-generation agent systems
- 💼 Background in Appian / low-code automation
- ♟️ Enjoy chess
- 🚀 Constantly leveling up toward stronger AI engineering skills
Building practical AI systems that are useful, modern, and ready for real-world use.