Daily Reading List – May 28, 2024 (#327)

Whew, what a Tuesday. I had an outstanding 3-day weekend with sunshine, friends, and baseball. I also (mostly) completed a fun coding project that I’ll blog about later this week. Today was a blur, but a lot got done. I think.

[article] A Great Sales Pitch Hinges on the Right Story. It doesn’t matter if you’re in sales, engineering, program management, or most any other role. Get good at storytelling!

[blog] Effective large language model adaptation for improved grounding. Some cool work from Google Research that looks at a new framework for adapting a base LLM to self-ground responses.

[blog] “The Business” is BS. If you’re in IT, or a tech consultant, don’t refer to a set of people as “the business.” It creates an unnecessary separation, and treats tech as a far-off service provider.

[blog] The Boring Product Manifesto. Making products shouldn’t be so dramatic. John says that we need more of the “good kind” of boring.

[blog] Using LLMs to Learn From YouTube. This seems like a complicated architecture, but it gets the job done.

[blog] Lazy Work, Good Work. Massively important point. Our most creative work, and the moments where we connect the dots, doesn’t happen in meetings. Get more thinking time.

[blog] Don’t Get Lost in the Metrics Maze: A Practical Guide to SLOs, SLIs, Error Budgets, and Toil. Here’s a brief, helpful take on some of the core ideas behind Site Reliability Engineering and focusing on the right dimensions when keeping a system online and healthy.

[blog] Grounding Gemini with Web Search results in LangChain4j. Read this for an excellent example of how to call an LLM and ground the results in a trustworthy source.

[blog] The future of foundation models is closed-source. Those building “open” models aren’t doing charity work, There’s other motives, and John encourages thinking about which models you’re betting on.

[blog] What if…slower wasn’t safer? Instead of “getting it right” by slowing down, maybe it’s smarter to make the inevitable process of making mistakes cheaper and faster? That’s the argument here.

[blog] The Future is Now: TuringBots Will Collapse the Software Development Life Cycle Siloes. I like the work that Forrester has done on AI dev assistants. Diego talks here about the changing SDLC.

##

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

Author: Richard Seroter

Richard Seroter is currently the Chief Evangelist at Google Cloud and leads the Developer Relations program. He’s also an instructor at Pluralsight, a frequent public speaker, the author of multiple books on software design and development, and a former InfoQ.com editor plus former 12-time Microsoft MVP for cloud. As Chief Evangelist at Google Cloud, Richard leads the team of developer advocates, developer engineers, outbound product managers, and technical writers who ensure that people find, use, and enjoy Google Cloud. Richard maintains a regularly updated blog on topics of architecture and solution design and can be found on Twitter as @rseroter.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.