#36 Tokens: The New Currency - with Tejas Chopra
How headroom is turning the world of tokens upside down My guest today is Tejas Chopra. Tejas is a software engineer, AI/ML infrastructure expert, and entrepreneur. He focuses extensively on the infrastructure behind AI and AI agents. With Headroom, he has created an open-source project designed to make LLM context more efficient, thereby reducing token consumption and AI costs. As a speaker and educator, he combines deep technical expertise with the ability to explain complex developments in AI in a way that’s both understandable and practical. Among other things, Tejas explains what Headroom is all about, how customers can benefit from it, and what it takes to work with Headroom. Together, we discuss how output quality is ensured when tokens aren’t used in the model. We’ll explore the technical and organizational aspects of Headroom, what it takes to get started with it, and how it can be used successfully. We’ll also talk about how Headroom is influencing the market and what we can expect from providers of large language models as Headroom turns their existing business models upside down. Other questions we address: What are the do’s and don’ts to start with? Where should one use Headroom? How easy is it to control a system like that? How do you handle the issue of control? When hiring new employees, should companies set aside specific funds for the use of tokens? Will an understanding of token efficiency and their application become a personal skill and part of the hiring process? What happens if you condense the context too much and how can the original information be preserved and conveyed? --- 👋🏼 Follow Tejas on LinkedIn: https://www.linkedin.com/in/chopratejas/ 🔗Check out Headroom: https://headroomlabs.ai 🔗 Check out Headroom on Github: https://github.com/headroomlabs-ai/headroom --- 🚨 Follow the podcast so you don't miss another episode ⭐️ I am happy about your 5-star rating