Aman Agrawal - Portfolio

About Me

Hi, I'm Aman Agrawal!

I always have a lot to write when it comes to machine learning and related topics! You'll mostly find me reading or sharing my learnings on AI-related subjects. I approach problems through a First Principles thinking methodology.

I have built projects in NLP and GenAI fields, with industry experience gained through internships in these domains. I believe in the concept of building in public, so you'll find me actively engaging on platforms like Medium, Twitter, and LinkedIn, constantly learning and sharing knowledge.

I write articles on Data Science and ML/AI topics on Medium. Feel free to explore my work here!
Connect via email: amannagrawall002@gmail.com

Experience

Pascal AI Labs

May 2025 - Present

Machine Learning Engineer Intern

Developed a comprehensive LLM evaluation platform enabling finance domain experts to conduct efficient prompt experimentation and systematic analysis. Built intuitive comparison tools and backend infrastructure, improving team efficiency and collaboration. Enhanced RAG accuracy by 11% for normal and 10% for advanced systems through research and implementation of system enhancements.

Job Twine

Dec 2024 - March 2025

Generative AI Intern

Developed the Playbook feature, an intelligent interview creation tool using OpenAI APIs and LangChain. Built scalable APIs with 60% improved customization. Leveraged RAG with PostgreSQL and Redis, reducing update times by 50%. Collaborated with CTO to drive 40% boost in user engagement and product quality.

Suryavanshi Ventures

Jan 2024 - July 2024

Artificial Intelligence Intern

Developed an AI-powered conversational training platform for 5-star hotel staff, simulating realistic customer interactions with real-time feedback. Leveraged OpenAI models, RAG-based search, and parallel processing to improve training accuracy by 50% and reduce latency by 30%.

Projects

GitHub Issues Semantic Search Engine

Built a robust data pipeline using GitHub REST API, implementing efficient caching and semantic search capabilities. Enhanced system performance through selective data extraction and streamlined retrieval processes. View Project →

Articles

I am passionate about writing and sharing knowledge through Medium articles, with over 15k views. I contribute as a writer to leading publications like Towards Data Science, Towards AI, and AI Advances, focusing on insights in data science, AI, and technology.

Towards Data Science

Bessel's Correction

Understanding why we divide by n−1 for sample variance is crucial in statistics. This method corrects for bias in estimating population variance from samples. Read more...

Towards AI

RAG for Knowledge-Intensive NLP Tasks

Retrieval-Augmented Generation enhances AI language models by integrating retrieval into the generation process. Discover the mechanics behind RAG. Read more...