Looking under the leaky abstractions of LLM

After a series of Fintech topics, we are back to discussing a tech topic. Today we are using LLMs to do a lot of work for us. Let's understand them from their fundamentals so we can better leverage them.

Speaker - Vedant Bhutani
I am a science enthusiast working currently as a software engineer at Postman. I did my engineering from IIT Gandhinagar (name dropping the college to gain some credibility 😅)

Talk details
In my talk I will try to break down the technology that powers LLMs from first principles, keeping the jargon to a minimum. To make an amazing AI product, one must go looking beyond the leaky abstractions that an LLM is. This talk is targeted for anyone who is interested in understanding the LLM technology better. I myself am not an expert on the area. I am an engineer who has been working with this technology a lot these days. I am an enthusiast when it comes to first principles thinking, and will try to bring my unique flavour to the talk.

Useful for Product Managers and Programmers working in Fintech and using LLMs.

Agenda
1.⁠ Talk and thought experiments on LLMs - Vedant (60m)
2.⁠ ⁠Networking (30m)