Beyond public repos, I have heavy experience building production-grade systems in corporate and startup settings:
- Designing and deploying on-prem LLM inference platforms with microservices (Docker, RabbitMQ, storage, APIs).
- Building enterprise-grade RAG systems (ingestion, embedding pipelines, hybrid retrieval, metadata filtering, efficient ranking) for large-scale documents.
- Fine-tuning LLMs (e.g. Qwen / LLaMA variants) using PEFT/SFT; quantization for optimized inference; distributed training on cloud/cluster setups.
- Implementing agentic AI workflows integrating external enterprise data sources (e.g. Salesforce) for contextual querying and automated reporting.
These experiences are part of my professional/freelance portfolio, which I’m happy to discuss or present privately — they reflect real-world impact even if not fully open-sourced.
📊 GitHub Stats:
- Passionate about building end-to-end AI & ML systems — from data preprocessing and embeddings to inference and deployment.
- Enjoy solving hard technical problems, especially around retrieval, embedding pipelines, scalability, distributed compute and RAG systems.
- Always experimenting, always learning — my public repos reflect the “sandbox” where I try out new ideas; my private / industry work reflects the “production” where I deliver at scale.
If you like what you see — feel free to explore my repos, drop a star, or reach out!