How to Build Terrible AI Systems with Jason Liu

​Jason is an independent consultant who uses his expertise in recommendation systems to help fast-growing startups build out their RAG applications. He was previously at Meta and Stitch Fix is also the creator of Instructor, Flight, and an ML and data science educator.

​In this live podcast recording, Hugo and Jason talk about how he approaches consulting companies across many industries, including construction and sales, in building production LLM apps, his playbook for getting ML and AI up and running to build and maintain such apps, and the future of tooling to do so.

​They’ll take an inverted thinking approach, envisaging all the failure modes that would result in building terrible AI systems, and then figure out how to avoid such pitfalls. They’ll talk about

• ​Jason’s LLM and RAG consulting playbook (see here for a sneak peek);

• ​How RAG is really RecSys in disguise;

• ​The most common mistakes people make in building AI systems;

• ​Things to consider when building MVPs;

• ​How to do the hardest part: data management;

• ​What we can learn tool and API-wise from frameworks such as the tidyverse and requests;

• ​Data literacy as the key, once again, but in a different light.

​

Time permitting, they may even talk about Jason as entrepreneur, how our Burnout Society incentivizes the creation of entrepreneurs, and how AI and data are implicated.