I build systems that think.
Then ship them.
Engineering LLM systems, inference pipelines, and production ML infrastructure. From model optimization to scalable deployment.
Engineering mindset, not buzzwords.
I work at the intersection of deep learning research and production systems. My focus is on making ML systems that actually work at scale — not just in notebooks, but in real infrastructure serving real traffic.
Core areas: LLM inference optimization, model quantization, transformer architectures, distributed training, and end-to-end ML pipeline design. I care about latency, throughput, and reliability as much as model accuracy.
When I write about AI, I write code. Every article on this site comes with implementation details, not just theory.
Things I've built.
HTML TriScope
RAG-powered system that answers queries by synthesizing information from three web sources simultaneously. Built with LangChain and deployed on Streamlit.
YouTube Summarizer
LLM-powered pipeline that extracts transcripts from YouTube videos and generates structured summaries. Handles long-form content with chunked processing.
Arxiv Paper Evaluator
Automated research paper assessment system. Parses Arxiv papers, evaluates methodology, novelty, and reproducibility using structured LLM analysis.
More coming soon
Working on inference optimization tooling, fine-tuning pipelines, and edge deployment systems.
Technical deep dives.
Quick Books: Key Learnings from Recent Reads
Distilled insights from books on technology, AI, and engineering — the golden takeaways worth remembering.
FrameworksFramework FastTrack
Quick refresher guides for data-related frameworks — from TensorFlow to PyTorch, simplified for rapid learning.
FundamentalsCross Entropy Loss
A comprehensive guide to understanding and implementing cross entropy in machine learning — from information theory to PyTorch.
FundamentalsElements of Neural Nets: Building Blocks of AI
An essential guide to the foundational components of modern neural networks — activations, cost functions, optimizers, and regularization.
Let's talk.
Building something interesting in AI? Have a question about one of my articles? Reach out.