Kunal Mangal

prof_pic.jpg

Hi! I’m an applied AI/ML engineer, specializing in NLP and document processing at scale. I’m based in Berkeley, CA.

Currently, I work as a a Senior Data Scientist Neo.Tax, where we automate enterprise accounting processes. As an early hire, I’ve had a chance to build many of the predictive systems that power the product from scratch. Some of my favorite projects include:

  • Engineering an agentic pipeline that mimics IRS auditor review
  • Building a custom evaluation framework for a clustering pipeline where standard methods (ground-truth labels, silhouette score) didn’t apply
  • Designing a hybrid retrieval pipeline to match Jira tickets to user-defined projects by semantic similarity

Previously, I was an academic economist researching labor markets in developing countries. You can read more about my research here.

Outside of work, I like to find side projects where I can put my engineering skills towards causes that I care about.

I’m always excited to meet other people working on similar problems—if that’s you, feel free to reach out!