What happens when the model CAN'T fix it? Interview w/ software engineer Landon Gray [Podcast #213]

Today Quincy Larson interviews Landon Gray. He's a package technologist who worked astatine agencies for years. Then he taught himself AI assisted package development. And now he's helping different devs do the same. Landon's celebrated for proving that RAG pipelines tin beryllium written successful Ruby and popularizing Ruby arsenic a connection for building instrumentality learning projects. He useful arsenic an AI Engineer astatine a endeavor package institution and runs a celebrated newsletter. We talk about: - How Large Language Models are conscionable the earthy fuel, and harnesses are the existent motor to get things done - Why building your master web is truthful adjuvant for uncovering clients and landing occupation interviews - Why Landon helped larboard Python instrumentality learning libraries to Ruby, and why he thinks that – now that AI is conscionable an API telephone distant – the Ruby ecosystem is better-positioned than ever. Support for this podcast comes from the 10,113 benignant folks who donate to our kindness each month. Join them and support our ngo astatine https://donate.freecodecamp.org Get a freeCodeCamp tshirt for $20 pinch free shipping anyplace successful the US: https://shop.freecodecamp.org Links from our discussion: - Landon's Substack newsletter: https://landongray.substack.com Community news section: 1. freeCodeCamp conscionable published a caller YouTube people that will thatch you beginner Front-end Development skills for illustration HTML, CSS, and JavaScript. You tin codification on astatine location and build a assortment of projects: your ain interactive quiz game, a rate converter app, and moreover a Trello-style kanban board. Along the measurement you'll study really to usage APIs and section retention to widen the functionality of these bite-sized apps. (12 hr YouTube course): https://www.freecodecamp.org/news/build-19-web-dev-projects-using-html-css-javascript/ 2. Learn really to decently trial your package and guarantee it doesn't break erstwhile you adhd caller features. Prolific freeCodeCamp coach Beau Carnes teaches this course. He'll present you to the Testing Pyramid and show you really to equilibrium accelerated portion tests against analyzable end-to-end personification journeys. You'll besides study really to automate immoderate of this testing utilizing an unfastened root room called Playwright and an LLM testing tool. (1 hr YouTube course): https://www.freecodecamp.org/news/software-testing-with-playwright/ 3. More and much apps are relying connected probabilistic LLM output alongside deterministic API calls. This makes life harder for devs who now request to guarantee that hallucinations don't flight to extremity users. freeCodeCamp conscionable published this precocious observability tutorial that will thatch you emerging champion practices and architectural patterns for dealing pinch this. (40 infinitesimal read): https://www.freecodecamp.org/news/build-end-to-end-llm-observability-in-fastapi-with-opentelemetry/ 4. Learn really to containerize your MLOps pipelines. This tutorial is the consequence of hard-won deployment wisdom. The writer spent 3 weeks debugging a Python room correction owed to dependency conflicts. His eventual answer: containerize full task pinch Docker. This tutorial will show you really to building your containers pinch multi-stage builds. You'll besides study really to group up research search pinch MLflow, versioning pinch DVC, GPU passthrough, and different precocious techniques. (40 infinitesimal read): https://www.freecodecamp.org/news/containerize-mlops-pipeline-from-training-to-serving/ 6. Today's opus of the week is 2006's Everybody by UK producers Basement Jaxx. If you're acquainted pinch their work, you cognize you're successful for a psychedelic yet silly romp. Between the spoons, bongos, and swooning chorus the opus feels for illustration it's held together pinch duct portion but it works. https://www.youtube.com/watch?v=OrMot81VE8g