Searching by intent (aka semantic search) is the antithesis of searching by keywords. Amazon’s semantic search solution is pretty amazing at how it leverages mega LLMs to (a) build knowledge graph to serve known queries, and (b) create smaller LLMs for generative responses to serve never-before-seen queries.
AI token usage in performance reviews? Seriously?!
Career growth in 2 sentences - ‘Your company will never tell you that you’ve gotten too comfortable because they benefit from your comfort. You have to be the one who notices it and breaks out of it.’
Introducing AI agents in a classic RAG search pipeline lends it the ability to think, reason, and decide whether the one-shot vector-search results are any good or require refinement. Agentic RAG might also be a good application to learn about agentic systems in general.
Intriguing use of LLM embeddings to link member data and post together for super-fast retrieval in social feeds. And, of course, the use of LLM as semantic search layer is 🆒
Didn’t know GitHub was in such dire waters. But TBH I like Copilot.
Certainty in uncertain times goes a long way in approaching touch situations with calm and clarity.
What not to prematurely optimize is one of the most important engineering decisions in system design. Like how Dropbox avoided generating preview thumbnails for all uploaded video/images based on their data-backed insight that most such files will never be viewed in a search result. Saved them both storage and compute costs.
Tap compare, what a clever way to migrate from one tech stack to another in production. Sounds like a beginner interviewer shadowing an experienced one in real interviews.
Development has ground to a halt. The thing is plagued by security issues. As fondly as I remember MySQL from my formative years, the sad reality is it stopped being a viable option soon after its Oracle acquisition. Why Sun, why did you sell yourself?