Joel Ross
@rosscode.com
Christian, father, husband.
Co-founded https://augusto.digital, https://tourneytopia.com, and https://payit2.com
Co-founded https://augusto.digital, https://tourneytopia.com, and https://payit2.com
This seems like the next step. I really enjoy NotebookLM, but sometimes it would be nice to have more LLM features, rather than just the RAG-like features it provides.
phandroid.com/2025/12/15/g...
phandroid.com/2025/12/15/g...
Google Is Connecting NotebookLM to Gemini, and Your Research Just Got Smarter - Phandroid
Google is testing NotebookLM integration with Gemini, letting the AI access your research notebooks directly for grounded memory.
phandroid.com
December 15, 2025 at 6:37 PM
This seems like the next step. I really enjoy NotebookLM, but sometimes it would be nice to have more LLM features, rather than just the RAG-like features it provides.
phandroid.com/2025/12/15/g...
phandroid.com/2025/12/15/g...
I've been wanting this for a while. We've attempted something similar through Slack, but the context around the conversation doesn't quite work right. Having a true group chat and being able to collaborate with ChatGPT involved will be nice.
www.testingcatalog.com/openai-readi...
www.testingcatalog.com/openai-readi...
OpenAI readies ChatGPT Group Chats with custom controls
OpenAI plans Group Chats for ChatGPT, letting teams now coordinate shared exchanges with custom prompts, potentially set for a December launch.
www.testingcatalog.com
November 18, 2025 at 4:24 PM
I've been wanting this for a while. We've attempted something similar through Slack, but the context around the conversation doesn't quite work right. Having a true group chat and being able to collaborate with ChatGPT involved will be nice.
www.testingcatalog.com/openai-readi...
www.testingcatalog.com/openai-readi...
Interesting take on AI's favorite character and everyone else's bane of existence—the em dash!
www.seangoedecke.com/em-dashes/
www.seangoedecke.com/em-dashes/
Why do AI models use so many em-dashes?
If you asked most people to name a defining feature of AI-generated writing, they’d probably say the em-dash — like this. Language models use em-dashes so much…
www.seangoedecke.com
November 4, 2025 at 6:08 PM
Interesting take on AI's favorite character and everyone else's bane of existence—the em dash!
www.seangoedecke.com/em-dashes/
www.seangoedecke.com/em-dashes/
Remember when blog posts just answered the question instead of pitching a newsletter or hiding behind a paywall?
October 13, 2025 at 4:40 PM
Remember when blog posts just answered the question instead of pitching a newsletter or hiding behind a paywall?
Python 3.14 was released. They're missing a golden opportunity to call it Pi-thon.
www.python.org/downloads/re...
www.python.org/downloads/re...
Python Release Python 3.14.0
The official home of the Python Programming Language
www.python.org
October 10, 2025 at 5:25 PM
Python 3.14 was released. They're missing a golden opportunity to call it Pi-thon.
www.python.org/downloads/re...
www.python.org/downloads/re...
"I don't know." 3 simple words, but it took a long time for me to learn to say that as a consultant. Now ChatGPT's evolving to realize the importance of that statement.
Now, it just needs to learn to follow up with "...but I'll find out."
www.talkandroid.com/515969-chatg...
Now, it just needs to learn to follow up with "...but I'll find out."
www.talkandroid.com/515969-chatg...
ChatGPT-5 now says “I don’t know” and that’s a game changer - Talk Android
In a world of chatbots that always seem to have an answer — even when they probably shouldn’t — something unexpected is happening. ChatGPT-5, the latest
www.talkandroid.com
October 1, 2025 at 4:33 PM
"I don't know." 3 simple words, but it took a long time for me to learn to say that as a consultant. Now ChatGPT's evolving to realize the importance of that statement.
Now, it just needs to learn to follow up with "...but I'll find out."
www.talkandroid.com/515969-chatg...
Now, it just needs to learn to follow up with "...but I'll find out."
www.talkandroid.com/515969-chatg...
I suppose if we're going to rely on agents to surf the web for us and make reservations, book travel, etc. we need a protocol that allows agents to make payments too, right?
cloud.google.com/blog/product...
cloud.google.com/blog/product...
Announcing Agent Payments Protocol (AP2) | Google Cloud Blog
Learn more about the Agent Payments Protocol (AP2), an open protocol that builds on A2A, Agent to Agent Protocol. AP2 was developed by Google with leading payments and technology companies to…
cloud.google.com
September 23, 2025 at 10:32 PM
I suppose if we're going to rely on agents to surf the web for us and make reservations, book travel, etc. we need a protocol that allows agents to make payments too, right?
cloud.google.com/blog/product...
cloud.google.com/blog/product...
This makes sense after thinking about it. Gen Z seems more tied to purpose than flexibility. That’s a good thing. Remote work can make it harder to feel that purpose.
sherwood.news/culture/gen-...
sherwood.news/culture/gen-...
Gen Z are less likely to favor fully remote working compared with millennials and boomers
Per a new Gallup survey, it turns out young workers do want to go back to the office... sometimes....
sherwood.news
September 19, 2025 at 11:43 PM
This makes sense after thinking about it. Gen Z seems more tied to purpose than flexibility. That’s a good thing. Remote work can make it harder to feel that purpose.
sherwood.news/culture/gen-...
sherwood.news/culture/gen-...
New Codex, less tokens. That was the first thing I noticed when running Codex - the token usage was significantly higher than doing the same task in Gemini or Claude Code. Hopefully this fixes that.
openai.com/index/introd...
#Codex #contextengineering
openai.com/index/introd...
#Codex #contextengineering
Introducing upgrades to Codex
Codex just got faster, more reliable, and better at real-time collaboration and tackling tasks independently anywhere you develop—whether via the terminal, IDE, web, or even your phone.
openai.com
September 18, 2025 at 5:52 PM
New Codex, less tokens. That was the first thing I noticed when running Codex - the token usage was significantly higher than doing the same task in Gemini or Claude Code. Hopefully this fixes that.
openai.com/index/introd...
#Codex #contextengineering
openai.com/index/introd...
#Codex #contextengineering
AI is changing a lot about what I do day to day, but leadership skills will always live on.
"...the heart of management remains timeless: setting purpose, building trust, helping people grow. "
lg.substack.com/p/managing-a...
"...the heart of management remains timeless: setting purpose, building trust, helping people grow. "
lg.substack.com/p/managing-a...
You're definitely going to be a manager now
Book launch! + Predictions for the manager in the era of AI
lg.substack.com
September 17, 2025 at 2:07 PM
AI is changing a lot about what I do day to day, but leadership skills will always live on.
"...the heart of management remains timeless: setting purpose, building trust, helping people grow. "
lg.substack.com/p/managing-a...
"...the heart of management remains timeless: setting purpose, building trust, helping people grow. "
lg.substack.com/p/managing-a...
Great guide to go from simple prompts and prompt engineering to moving to context engineering. We all start someplace, but if don't keep learning, our skills will get stagnant - faster with AI.
pub.towardsai.net/from-prompts-to-context-the-ai-revolution-thats-changing-everything-131468919f5b
pub.towardsai.net/from-prompts-to-context-the-ai-revolution-thats-changing-everything-131468919f5b
From Prompts to Context: The AI Revolution That’s Changing Everything 🚀
Why Context Engineering is the Secret Sauce That Transforms AI from Clever Chatbot to Reliable Business Partner
pub.towardsai.net
September 16, 2025 at 8:01 PM
Great guide to go from simple prompts and prompt engineering to moving to context engineering. We all start someplace, but if don't keep learning, our skills will get stagnant - faster with AI.
pub.towardsai.net/from-prompts-to-context-the-ai-revolution-thats-changing-everything-131468919f5b
pub.towardsai.net/from-prompts-to-context-the-ai-revolution-thats-changing-everything-131468919f5b
Got to listen to Steve Wozniak speak today at the kickoff event for Tech Week GR. He only had an hour, but I could have listened to the stories of his pranks all night.
September 16, 2025 at 2:09 AM
Got to listen to Steve Wozniak speak today at the kickoff event for Tech Week GR. He only had an hour, but I could have listened to the stories of his pranks all night.
This is a good reminder of where AI code stands (at least for now) and what to do about it: Don't Freak Out
www.honeycomb.io/blog/disposa...
#vibecoding
www.honeycomb.io/blog/disposa...
#vibecoding
Disposable Code Is Here to Stay, but Durable Code Is What Runs the World
Every day I seem to run into yet another post with someone solemnly opining that “writing code has never been the hardest part of software engineering. And hey, that’s smashing.
www.honeycomb.io
September 15, 2025 at 4:15 PM
This is a good reminder of where AI code stands (at least for now) and what to do about it: Don't Freak Out
www.honeycomb.io/blog/disposa...
#vibecoding
www.honeycomb.io/blog/disposa...
#vibecoding
Lovable at $1.8B? Chump change, I guess. Replit hits $3B. AI dollars aren't real. #AI #Replit
www.reuters.com/business/ai-...
www.reuters.com/business/ai-...
www.reuters.com
September 14, 2025 at 8:30 PM
Lovable at $1.8B? Chump change, I guess. Replit hits $3B. AI dollars aren't real. #AI #Replit
www.reuters.com/business/ai-...
www.reuters.com/business/ai-...
This is a great prompting guide. I've picked up most of this slowly over time. Having this up-front would have been helpful. My favorite is right at the end - having the AI write the prompt for you!
medium.com/age-of-awareness/the-3-level-prompting-system-that-makes-ai-insanely-useful-c37f9e64d547
medium.com/age-of-awareness/the-3-level-prompting-system-that-makes-ai-insanely-useful-c37f9e64d547
The 3-Level Prompting System That Makes AI Insanely Useful
Learn the science-backed, high-impact way to design prompts that think, create, and solve like a human partner.
medium.com
September 13, 2025 at 8:00 PM
This is a great prompting guide. I've picked up most of this slowly over time. Having this up-front would have been helpful. My favorite is right at the end - having the AI write the prompt for you!
medium.com/age-of-awareness/the-3-level-prompting-system-that-makes-ai-insanely-useful-c37f9e64d547
medium.com/age-of-awareness/the-3-level-prompting-system-that-makes-ai-insanely-useful-c37f9e64d547
For all the noise and fear around AI, there's a lot of very cool advances coming, especially in healthcare. Time to get new stethescopes!
nypost.com/2025/09/05/t...
nypost.com/2025/09/05/t...
AI stethoscope can detect dangerous heart conditions in seconds
Researchers at Imperial College London and Imperial College Healthcare NHS Trust discovered that an AI stethoscope can detect heart failure at an early stage.
nypost.com
September 12, 2025 at 1:45 PM
For all the noise and fear around AI, there's a lot of very cool advances coming, especially in healthcare. Time to get new stethescopes!
nypost.com/2025/09/05/t...
nypost.com/2025/09/05/t...
Training AI models is over my head, but this helped me understand one part: current eval methods reward confident guesses over "I don't know."
Consultants often start that way too. It takes experience to realize "I don't know" is valid.
openai.com/index/why-la...
Consultants often start that way too. It takes experience to realize "I don't know" is valid.
openai.com/index/why-la...
Why language models hallucinate
OpenAI’s new research explains why language models hallucinate. The findings show how improved evaluations can enhance AI reliability, honesty, and safety.
openai.com
September 10, 2025 at 3:20 PM
Training AI models is over my head, but this helped me understand one part: current eval methods reward confident guesses over "I don't know."
Consultants often start that way too. It takes experience to realize "I don't know" is valid.
openai.com/index/why-la...
Consultants often start that way too. It takes experience to realize "I don't know" is valid.
openai.com/index/why-la...
If you've read or seen Moneyball, It's fitting that a baseball team in Oakland would be pushing boundaries.
Oakland Ballers to use artificial intelligence to manage Saturday home game against Great Falls
The playoff-bound Oakland Ballers of the independent Pioneer League are turning to artificial intelligence to manage all aspects of Saturday’s home game against the Great Falls Voyagers at Raimondi…
apnews.com
September 8, 2025 at 2:55 PM
If you've read or seen Moneyball, It's fitting that a baseball team in Oakland would be pushing boundaries.
I believe this is what Sam Altman calls "the fast fashion era of software" right? I'm certainly enjoying being able to quickly write an app that gets me something now rather than waiting weeks, even if it gets set aside later. Plus, I feel like a software engineer again!
a16z.com/disposable-s...
a16z.com/disposable-s...
Disposable Software | Andreessen Horowitz
When software was expensive and hard to build, we mostly made it for serious things and expected it to last (like payroll systems, tax tools, ERP software and consumer networks). Every line of code…
a16z.com
September 4, 2025 at 1:45 PM
I believe this is what Sam Altman calls "the fast fashion era of software" right? I'm certainly enjoying being able to quickly write an app that gets me something now rather than waiting weeks, even if it gets set aside later. Plus, I feel like a software engineer again!
a16z.com/disposable-s...
a16z.com/disposable-s...
AI is making strange bedfellows. This is what happens when you're up against a dominant player and falling behind.
www.bloomberg.com/news/article...
www.bloomberg.com/news/article...
www.bloomberg.com
August 28, 2025 at 1:15 PM
AI is making strange bedfellows. This is what happens when you're up against a dominant player and falling behind.
www.bloomberg.com/news/article...
www.bloomberg.com/news/article...
We've been experimenting using NotebookLM as a knowledge source for projects. Most of our documentation is in Google Docs, so this is interesting to me. It's fun to see this evolve. I was afraid this would end up as another abandoned Google tool.
www.testingcatalog.com/notebooklm-t...
www.testingcatalog.com/notebooklm-t...
NotebookLM to pull sources from web via Deep Research soon
What we know so far: The Deep Research feature from Gemini is coming to NotebookLM. Likely with a possibility to pull sources from Google Drive as well.
www.testingcatalog.com
August 22, 2025 at 6:45 PM
We've been experimenting using NotebookLM as a knowledge source for projects. Most of our documentation is in Google Docs, so this is interesting to me. It's fun to see this evolve. I was afraid this would end up as another abandoned Google tool.
www.testingcatalog.com/notebooklm-t...
www.testingcatalog.com/notebooklm-t...