Daily Reading List – September 26, 2024 (#406)

We’re doing some performance reviews and promotion assessments right now at work. I’m learning to give more specific feedback and lean into direct conversations. But I feel very fortunate to work with such a high caliber team that continues to perform so well.

[blog] Some advice and good practices when integrating an LLM in your application. Excellent post. Read this before going too far down a path with generative AI models.

[blog] Why I still blog after 15 years. Loved this. Yesterday, I published my 1000th post on this blog, and have been blogging overall for more than 19 years. This author’s reasons for continuing match mine.

[doc] Demystifying AI Risk: An Actionable Framework Aligning Business Needs to Risks and Mitigations. There’s useful content here for folks who want to learn about the unique risks to mitigate when you bring AI into your environment.

[article] AI Can (Mostly) Outperform Human CEOs. Any role that thinks its immune from the impact of AI, is mistaken.

[article] Russ Cox’s Next Act: AI-Powered Help Agents for Open Source Projects. Russ was a long-time tech lead for the Go language, and is now doing some cool work to make OSS maintenance easier.

[article] Meta’s Llama AI models now support images, too. Another day, another exciting model drops. The Meta folks are now in the multimodal game where they can handle more than just text input. Read Meta’s post too.

[blog] Meta’s Llama 3.2 is now available on Google Cloud. And the vendors are getting good at making new models available almost instantly. We offer a few low-touch ways to try it out.

[blog] NotebookLM adds audio and YouTube support, plus easier sharing of Audio Overviews. I’d bet that this service is about to breakthrough in a big way. Now with fresh audio input options and audio sharing. Create amazing notebooks out of piles of source material.

[article] Semantic Router and Its Role in Designing Agentic Workflows. If you’re deploying more than one model (or editions within a family of models), this is a pattern you should know about.

Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below: