
Vanishing Gradients Episode 24: LLM and GenAI Accessibility
4 snips
Feb 27, 2024 Hugo and Johno discuss the evolution of tooling and accessibility in AI over the past decade, highlighting advancements in using big models from Hugging Face and hi-res satellite data. They delve into the Generative AI mindset, democratizing deep learning with fast.ai, and the importance of UX in generative AI applications. The discussion also covers the skill set needed to be an LLM and AI guru, as well as efforts at answer.ai to democratize LLMs and foundation models.
AI Snips
Chapters
Transcript
Episode notes
11-Minute Image+Quote Prototype With ChatGPT
- Johno demoed building a simple app by asking ChatGPT to fetch an Unsplash photo, generate a quote, and overlay text on the image.
- The whole glue took about 11 minutes using natural-language prompts and code suggestions.
Use Notebooks To Glue And Test Building Blocks
- Break complex pipelines into building blocks and test each piece in a notebook before end-to-end runs.
- Use transcription, retrieval, and LLM summarization as modular steps to automate tasks like YouTube chaptering.
Research To Tooling Cycle Has Compressed
- Cutting-edge research moves to usable services faster now, shrinking the time from paper to deployable tool.
- The ecosystem keeps iterating rapidly, knocking rough edges off weeks or months after release.
