RAG to Riches — cover art: a composed red fox in a claret waistcoat sips tea at her desk, paw resting on a small brass spray-canister, while pinstriped weasels and rats lean out of the flanking topiary brandishing cloth money-sacks. 😄

2023 · Sourcegraph · Essay

“If you are an ML engineer, then you have by this time already figured out that you need to carry bear spray around to fend off recruiters.”
— From RAG to Riches, December 2023
Read the essay

© 2023 Steve Yegge. Originally published at Sourcegraph.

AI Notes

December 2023, end-of-year Sourcegraph piece framed as a news roundup — Sam Altman's Uno Reverse moment at OpenAI is the lead, and the now-famous Bear Spray Engineers riff is the centrepiece: ML talent is so contested that any engineer who wants a job needs bear spray to fend off the recruiters chasing them. A recurring deadpan kicker punctuates each imagined leadership exchange, and the catnip.ai unicorn and a Korean luxury-car interview anecdote round out the comic mode at its peak. Under the jokes, the real news is that Cody has gone GA and Steve is making the case that the differentiator is retrieval — the augmented "G" in RAG. Cody's context has "graduated from 'hey we have vector embeddings' to 'hey we have a whole squad of engines,'" each providing a different lens on the codebase. There's a side jab at Completion Acceptance Rate (CAR) as a terrible proxy for developer productivity, capped by a deadpan F1-racing metaphor.

A snapshot of where Sourcegraph thought the industry sat at the end of 2023: ML hiring is bonkers, models are good enough to wire up, and the next year's differentiation will be fought on context retrieval.

Related listings

Where it was argued