One day, I will write a nuanced post about using large language models in production. It will be better than all the nuanced posts published so far about using large language models in production. Until then, here are some brief lessons learnt from building and operating LLM APIs in a production environment for the last 6 months.

During the first 2-3 years of my career, I used to think that enterprise software can get by with average UX (user experience). I don’t believe that anymore. It depends a lot on the type of work the end-user does. The extent of involvement of managers and executives also plays a part.

In early June 2020, I decided to learn about the current state of NLP and correspondingly, role of (narrow) AI. My usual approach to understanding a topic is bottom-up. Understand the fundamentals thoroughly before starting out with a project. Constrained by time and inspired by a pedagogy promoted by the likes of Fast AI, I decided to go project-first instead.

Note: In addition to experience as an early startup employee, the below content has been derived from my time at the Entrepreneur First program.

It had been less than a week since I had entered college. I was standing in front of a crowded computer lab full of starry-eyed freshers. A senior (final-year student) drew a curve on the blackboard. It was either a parabola or a line. Truth is I don’t really remember. Over the next 30 minutes, I listened to the senior to explain how computer science is deeply intertwined with maths. I was surprised to learn that even those not from a computer science program could benefit from learning to code.

Note: Below content is informed by online research, reading and interviewing law practitioners. Resources can be found at end of the article.