About
I'm a Computer Science graduate student at the University of Kansas and a Software Engineer Intern on the AI/ML infrastructure team at Flyover Capital. I work where software engineering meets machine learning and data, building systems that retrieve, extract, and reason over messy real-world information.
At Flyover Capital I build hybrid semantic retrieval over knowledge graphs (ChromaDB and Neo4j) and LLM-powered data pipelines. Before that, I worked on information extraction from scanned medical documents at Smart Data Solutions, and built LLM and retrieval-based systems at Naamche. My interests span data engineering, distributed systems, NLP, and applied machine learning, and I like teaching what I learn along the way.
On AI
I firmly believe Artificial General Intelligence is possible. To those who doubt it: think about how human intelligence came to be. Isn't it the product of evolution? Nature, through randomness, created something capable of thought, reflection, and consciousness.
Now consider this: if randomness can produce intelligence, could intentionality and design be any less powerful? We are, after all, just clusters of atoms. If nature could generate intelligence from chance, surely we can consciously create it too. The real question is not whether we can create intelligence, but what we choose to do with it.
I'm optimistic about a future with AI, and I believe what we choose to do with it will be for the betterment of society. I'm optimistic about the job market, too. Like Jensen Huang, Nvidia's CEO, I don't see AI as simply replacing jobs; it makes people more productive, and that productivity lets companies save and reinvest into new ideas and new companies, creating new opportunities and, over time, a positive impact on society.
News
- Aug 2026Started as a Graduate Research Assistant at the University of Kansas, working on applied machine learning research.
- May 2026Joined Flyover Capital as a Software Engineer Intern.
- Jan 2026TA'd EECS 140 (Digital Logic Design) at the University of Kansas.
- Aug 2025Began my MS in Computer Science at the University of Kansas, and TA'd EECS 348 (Software Engineering).
- Jul 2025Wrapped up 1.5 years as a Machine Learning Engineer at Smart Data Solutions.
- Jan 2025Our paper on structured information extraction from Nepali scanned documents was presented at CHiPSAL @ COLING 2025, Abu Dhabi.
- 2024EaseAnnotate, a document-AI tool I helped build, launched as a SaaS and was backed by Microsoft for Startups.
Publications
COLING
2025
Structured Information Extraction from Nepali Scanned Documents using Layout Transformer and LLMs
Experience
Graduate Research Assistant
Applied machine learning research, building deep learning models for large scientific datasets. This work will form the basis of my MS thesis.
Software Engineer Intern (AI / ML Infrastructure)
Built a hybrid semantic retrieval system over 50,000+ entities, combining vector embeddings (ChromaDB) with a graph database (Neo4j) for relationship-aware search across the firm's network and CRM. Built LLM-powered pipelines for information extraction, entity enrichment, and knowledge-graph construction over 20,000+ documents and records, plus internal RAG tooling for natural-language querying.
Built MCP servers and tools that bring the firm's data into each team member's Claude, integrations with Airtable and Affinity, and a custom search agent. Also automated an events pipeline that reads invitation emails, extracts structured event data with an LLM, de-duplicates against Airtable, and posts to Slack.
Graduate Teaching Assistant
Software Engineering (EECS 348, Fall 2025): led weekly labs for 30+ undergraduates on C/C++, Git, Docker, and system design, and graded assignments and labs. Digital Logic Design (EECS 140, Spring 2026): ran hardware labs on FPGA design with VHDL on the Basys3 board, including soldering.
Machine Learning Engineer
Built document AI for healthcare, processing 5,000+ documents a day. Fine-tuned a Qwen multimodal model to extract structured fields from insurance claims (about 90% accuracy with spelling correction), and combined regex with ML for high-accuracy extraction of vital fields from electronic medical records.
Built document classifiers (a combined NLP and computer-vision model for text-plus-layout documents, and a Random Forest over word tokens), a CNN X-ray classifier at about 99% accuracy, and a RAG chatbot grounded in patient records. Worked in Python and Java and owned the services end to end, from testing through production monitoring. The automation cut manual review from three reviewers to two plus AI.
Machine Learning / Backend Engineer
Built production RAG systems and AI agents with LangChain for domain-specific Q&A, using embedding retrieval over vector databases. Engineered data ingestion and indexing with Scrapy and AWS SageMaker, shipped FastAPI services with evaluation and guardrails, and built Metabase dashboards.
Machine Learning Intern
Worked on NLP with RNNs and transformer-based attention models, and built conversational agents with the Rasa framework.
Technical Coordinator
Built and launched the online certification system, managed Azure infrastructure, and led a 10-day fellowship for 120+ students.
Education
MS, Computer Science
Machine Learning · Deep Reinforcement Learning · Analysis of Algorithms · Advanced Data Science · Information Retrieval · Static Program Analysis
BE, Computer Engineering
Linear Algebra · Probability & Statistics · Calculus I–III · Numerical Methods · Discrete Mathematics · Artificial Intelligence · Data Science · Operating Systems · Database Management Systems
Tutorials
Small, animated explainers that build the fundamentals from the ground up. I started them for my younger brother, and I keep making them because I genuinely love it. Each was built with AI as a hands-on collaborator, from drafting the explanations to generating the interactive visualizations.
Writing
Loading posts…