Pratik Bhavsar

Full Stack AI Engineer

Bay Bridge @ SF

Hello! I’m an AI engineer at Galileo, with a strong track record of building AI systems. My passion is turning cutting-edge research into real-world impact and I thrive on tackling ambitious AI challenges from prototype to production.

• Startup experience: Designed and implemented mission-critical search and AI at Taskhuman and Enterpret. I am currently leading open-source AI agent evaluation. My RAG evaluation research was featured in Andrew NG's newsletter.

• Enterprise leadership: Launched end-to-end AI initiatives from the ground up at Morningstar.

• Public presence

  • Interview with Latent Space podcast

  • 23k followers on LinkedIn

  • 3k followers on Twitter

• Open source:

  • Agent Leaderboard: Framework for evaluating LLMs based enterprise support agents

  • BRAG - a series of high-performance open-source LLMs for RAG.

• Recognition: Named one of the Top AI Developers to Watch in 2023.

• Community: Built Max - a research assistant for my Discord community Maxpool

• Hacking:

  • 1st prize in every company-hosted hackathon - Galileo, Morningstar, Enterpret

  • Placed in the top 20 nationally in Microsoft’s Search Ranking challenge.

  • Ranked among the top 20 teams in the SEMEVAL competition.

• My books on GenAI:

  • Mastering Multi-agents

  • Mastering Agents

  • Mastering RAG

  • Mastering LLM-as-a-Judge

I share tech and business insights on my Substack newsletter - pakodas.substack.com.

Much of my content is categorised at blogs.pratik.ai.

I have worked on a variety of domains and projects.

• Agent evaluation

• RAG Dataset, Evaluation & Training

• Few shot hyper classification

• Semantic search (multi-stage)

• Elasticsearch - search and recommendation engine

• Algorithmic trading with ML

Looking forward to connect with you on Twitter and LinkedIn

Here are some of my old hackathon stories: https://www.quora.com/What-happens-at-a-hackathon/answer/Pratik-Bhavsar