Real World AI Integration
summary
Teams benchmark LLMs specifically for Clojure code generation using an open‑source benchmark (added Clojure support by Alex Miller) and find Clojure scores slightly lower than Python/JS but improving; engineers are advised to play with models for ~50 minutes and note Hugging Face hosts over a million models. A tiny, compliant local proxy called Roxy routes requests and works in production. Socratic prompting and roleplay improve design thinking, and an OpenHands agent demo loads the Iris CSV, uses tablecloth, and produces a working summary.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 3:06What is an ML
- 8:08ML Models
- 13:45We dont know
- 14:06Build flexible tools
- 15:34Create a community
- 16:14Can LMS help us think
- 17:24Socratic method
- 18:08Roleplaying
- 19:35Live Demo
related talks
51mIntelligent Applications with Spring AI
30mBuilding Reliable AI Agents: An Open Source Approach To Evaluation and Observability
21mLLMs + Clojure = Who needs frameworks?
41mA Clojure fusion of symbolic and data driven AI
32mScaling Generative AI: Building Production-Ready LLM Applications
25mThree steps for OCaml to crest the AI humps
16mInterview of Josh Long
51mInternal AI solutions -from hype to production