Jusdino
@jusdino.frahm.social
Cloud software engineering leader in Colorado
https://www.linkedin.com/in/justinfrahm/
https://www.linkedin.com/in/justinfrahm/
WSJ and others say 100-hour work weeks are back, for the AI race, at least. I've done that before and it can be exhilarating for a little while. Then life passes you by.
Taking ownership of your work is great, but don't do this to yourselves. You also need to take ownership of your LIFE.
Taking ownership of your work is great, but don't do this to yourselves. You also need to take ownership of your LIFE.
October 25, 2025 at 10:48 PM
WSJ and others say 100-hour work weeks are back, for the AI race, at least. I've done that before and it can be exhilarating for a little while. Then life passes you by.
Taking ownership of your work is great, but don't do this to yourselves. You also need to take ownership of your LIFE.
Taking ownership of your work is great, but don't do this to yourselves. You also need to take ownership of your LIFE.
"Leadership isn’t just behavioral; it’s neurological. In a world saturated with inputs, the leaders who thrive won’t be those who try to think faster. They’ll be those who design better neural environments—working with the brain, not against it."
Worth a closer look.
hbr.org/2025/10/stop...
Worth a closer look.
hbr.org/2025/10/stop...
Stop Overloading the Wrong Part of Your Brain at Work
Most leaders rely heavily on one key region of the brain: the prefrontal cortex (PFC), which is responsible for high-order functions like focus, planning, self-regulation, and decision-making. But the...
hbr.org
October 18, 2025 at 4:50 PM
"Leadership isn’t just behavioral; it’s neurological. In a world saturated with inputs, the leaders who thrive won’t be those who try to think faster. They’ll be those who design better neural environments—working with the brain, not against it."
Worth a closer look.
hbr.org/2025/10/stop...
Worth a closer look.
hbr.org/2025/10/stop...
Ok, Cursor IDE's 'Plan Mode' is a flop. It makes a nice plan document before implementing, then immediately forgets the whole plan as soon as you start giving it feedback on it's implementation 🤦♂️
October 13, 2025 at 11:04 PM
Ok, Cursor IDE's 'Plan Mode' is a flop. It makes a nice plan document before implementing, then immediately forgets the whole plan as soon as you start giving it feedback on it's implementation 🤦♂️
Watch out AWS! Cursor just stole Kiro's one exceptional feature:
October 9, 2025 at 8:17 PM
Watch out AWS! Cursor just stole Kiro's one exceptional feature:
"Simulations showed the technology reduced energy consumption by 27.7% compared to traditional neural networks, while also demonstrating resilience by reorganizing after damage."
Wow, a self-reorganizing chip? That's amazing. I hope to see it in real products.
neurosciencenews.com/ai-chip-lear...
Wow, a self-reorganizing chip? That's amazing. I hope to see it in real products.
neurosciencenews.com/ai-chip-lear...
October 3, 2025 at 7:24 PM
"Simulations showed the technology reduced energy consumption by 27.7% compared to traditional neural networks, while also demonstrating resilience by reorganizing after damage."
Wow, a self-reorganizing chip? That's amazing. I hope to see it in real products.
neurosciencenews.com/ai-chip-lear...
Wow, a self-reorganizing chip? That's amazing. I hope to see it in real products.
neurosciencenews.com/ai-chip-lear...
@quinnypig.com , what do start-ups use when they're finally too fed up with AWS Cognito?
August 28, 2025 at 3:13 PM
@quinnypig.com , what do start-ups use when they're finally too fed up with AWS Cognito?
My go-to email security PSA, "Never send anything in email that you wouldn't be comfortable mailing on a postcard" seems a bit weak, now:
www.freep.com/story/news/h...
www.freep.com/story/news/h...
Michigan Medicine sends postcards without envelopes, exposing personal data of 1,015
Protected health information of 1,015 people was exposed when recruiters for a medical study at Michigan Medicine sent postcards without envelopes.
www.freep.com
August 15, 2025 at 3:49 PM
My go-to email security PSA, "Never send anything in email that you wouldn't be comfortable mailing on a postcard" seems a bit weak, now:
www.freep.com/story/news/h...
www.freep.com/story/news/h...
Who else is feeling the AI pinch? How would you adapt?
“The unfortunate thing right now, specifically for recent college grads, is those positions that are most likely to be automated are the entry-level positions that they would be seeking,”
www.nytimes.com/2025/08/10/t...
“The unfortunate thing right now, specifically for recent college grads, is those positions that are most likely to be automated are the entry-level positions that they would be seeking,”
www.nytimes.com/2025/08/10/t...
Goodbye, $165,000 Tech Jobs. Student Coders Seek Work at Chipotle.
www.nytimes.com
August 10, 2025 at 5:03 PM
Who else is feeling the AI pinch? How would you adapt?
“The unfortunate thing right now, specifically for recent college grads, is those positions that are most likely to be automated are the entry-level positions that they would be seeking,”
www.nytimes.com/2025/08/10/t...
“The unfortunate thing right now, specifically for recent college grads, is those positions that are most likely to be automated are the entry-level positions that they would be seeking,”
www.nytimes.com/2025/08/10/t...
I miss AWS Bedrock Studio. It was really nifty in 2024. It's now been like 6 months since they smashed it into AWS Sagemaker 'Unified' Studio and the UX for tinkering on Gen AI apps is still terrible. So disappointed.
August 5, 2025 at 2:20 AM
I miss AWS Bedrock Studio. It was really nifty in 2024. It's now been like 6 months since they smashed it into AWS Sagemaker 'Unified' Studio and the UX for tinkering on Gen AI apps is still terrible. So disappointed.
Anybody else ready to unleash the hounds (and cash) with a whole team of AI developers via Claude Subagents touching your production software?
medium.com/@max.petruse...
medium.com/@max.petruse...
I Spent 30 Days Building an AI Development Team. Here’s What Actually Happened.
The unvarnished truth about Claude Code’s subagents — including the failures, surprises, and why this changes everything
medium.com
August 4, 2025 at 3:19 PM
Anybody else ready to unleash the hounds (and cash) with a whole team of AI developers via Claude Subagents touching your production software?
medium.com/@max.petruse...
medium.com/@max.petruse...
"Why do employees secretly use ChatGPT while ignoring approved tools? Why do groups that could benefit most from AI augmentation adopt at the lowest rates? Why do companies see limited ROI despite massive AI investments?"
Fascinating read with useful insights:
hbr.org/2025/08/rese...
Fascinating read with useful insights:
hbr.org/2025/08/rese...
Research: The Hidden Penalty of Using AI at Work
Researchers conducted an experiment with 1,026 engineers in which participants evaluated a Python code snippet that was purportedly written by another engineer, either with or without AI assistance. T...
hbr.org
August 3, 2025 at 1:18 AM
"Why do employees secretly use ChatGPT while ignoring approved tools? Why do groups that could benefit most from AI augmentation adopt at the lowest rates? Why do companies see limited ROI despite massive AI investments?"
Fascinating read with useful insights:
hbr.org/2025/08/rese...
Fascinating read with useful insights:
hbr.org/2025/08/rese...
Tired of giving your IDE AI agent repetitive feedback for its code responses? Ask it to maintain a `CONTRIBUTING.md` based on your feedback and reference it in future requests.
July 24, 2025 at 1:05 AM
Tired of giving your IDE AI agent repetitive feedback for its code responses? Ask it to maintain a `CONTRIBUTING.md` based on your feedback and reference it in future requests.
Reposted by Jusdino
Amazon Q shipped a feature where a rando hacker told it to run aws iam delete-user, and AWS said “Sure thing, pal!”
They caught it only because a journalist asked.
This isn’t “move fast and break things," it's “move fast and let strangers write your roadmap.”
www.lastweekinaws.com/blog/amazon-...
They caught it only because a journalist asked.
This isn’t “move fast and break things," it's “move fast and let strangers write your roadmap.”
www.lastweekinaws.com/blog/amazon-...
Amazon Q: Now with Helpful AI-Powered Self-Destruct Capabilities - Last Week in AWS Blog
Today 404Media released a truly stunning report that almost beggars belief. To break it down into its simplest form: A hacker submitted a PR. It got merged. It told Amazon Q to nuke your computer and ...
www.lastweekinaws.com
July 23, 2025 at 6:14 PM
Amazon Q shipped a feature where a rando hacker told it to run aws iam delete-user, and AWS said “Sure thing, pal!”
They caught it only because a journalist asked.
This isn’t “move fast and break things," it's “move fast and let strangers write your roadmap.”
www.lastweekinaws.com/blog/amazon-...
They caught it only because a journalist asked.
This isn’t “move fast and break things," it's “move fast and let strangers write your roadmap.”
www.lastweekinaws.com/blog/amazon-...
While [Cursor IDE users increasingly feel that their paid subscriptions get them less and less](forum.cursor.com/t/frustrated...), the AWS [Kiro IDE](kiro.dev) launch sees such rapid adoption, they have to wait-list the download!
Amazon, throw me a bone, I was on vacation when you launched! 🥺
Amazon, throw me a bone, I was on vacation when you launched! 🥺
July 22, 2025 at 9:53 PM
While [Cursor IDE users increasingly feel that their paid subscriptions get them less and less](forum.cursor.com/t/frustrated...), the AWS [Kiro IDE](kiro.dev) launch sees such rapid adoption, they have to wait-list the download!
Amazon, throw me a bone, I was on vacation when you launched! 🥺
Amazon, throw me a bone, I was on vacation when you launched! 🥺
When using an AI IDE for software engineering, right-sizing your requested changes is a key skill. Too small and you lose efficiency in feedback cycles, too big and the model can get distracted, or waste a bunch of time with an unacceptable approach. A good prompt with the right size balances these.
July 20, 2025 at 4:47 AM
When using an AI IDE for software engineering, right-sizing your requested changes is a key skill. Too small and you lose efficiency in feedback cycles, too big and the model can get distracted, or waste a bunch of time with an unacceptable approach. A good prompt with the right size balances these.
Reposted by Jusdino
ICYMI: Systemic engineering bottlenecks won’t disappear by just throwing AI at them. You’ve got to fix the roadblocks to realize the full value of AI tools.
Throwing AI at Developers Won’t Fix Their Problems
Systemic engineering bottlenecks won’t disappear by just throwing AI at them. You’ve got to fix the roadblocks to realize the full value of AI tools.
bit.ly
July 17, 2025 at 10:30 PM
ICYMI: Systemic engineering bottlenecks won’t disappear by just throwing AI at them. You’ve got to fix the roadblocks to realize the full value of AI tools.
So, if I want to show up on this particular feed, I only need to say one of the magic keywords? Let's find out: DevOps. <.< >.>
July 17, 2025 at 6:48 AM
So, if I want to show up on this particular feed, I only need to say one of the magic keywords? Let's find out: DevOps. <.< >.>
AWS announced a new AI IDE, built on VS Code. Its dev workflow design looks a lot like what I found works well in Cursor (what I was leading up to with my other posts) only deliberately built in from the start. I'm eager to check it out!
kiro.dev
kiro.dev
Kiro
The AI IDE for prototype to production
kiro.dev
July 14, 2025 at 8:15 PM
AWS announced a new AI IDE, built on VS Code. Its dev workflow design looks a lot like what I found works well in Cursor (what I was leading up to with my other posts) only deliberately built in from the start. I'm eager to check it out!
kiro.dev
kiro.dev
"AI makes it easier to produce more code. And if we’re not taking care to make sure that code is high quality, that our pipelines are ready ..., we can do some lasting damage."
Leveraging AI while also maintaining quality is an underdeveloped skill right now.
thenewstack.io/how-to-measu...
Leveraging AI while also maintaining quality is an underdeveloped skill right now.
thenewstack.io/how-to-measu...
How to Measure the ROI of AI Coding Assistants
Pressure to adopt AI does not mean success. An AI strategy hinges on the ability to measure the impact of AI agents and coding assistants.
thenewstack.io
July 13, 2025 at 5:07 PM
"AI makes it easier to produce more code. And if we’re not taking care to make sure that code is high quality, that our pipelines are ready ..., we can do some lasting damage."
Leveraging AI while also maintaining quality is an underdeveloped skill right now.
thenewstack.io/how-to-measu...
Leveraging AI while also maintaining quality is an underdeveloped skill right now.
thenewstack.io/how-to-measu...
Cross-domain thinking - a key to innovation:
"The capacity to see connections across different domains of knowledge is extremely valuable for kids today, as artificial intelligence becomes better at replacing traditional deep understanding of content."
www.psychologytoday.com/us/blog/the-...
"The capacity to see connections across different domains of knowledge is extremely valuable for kids today, as artificial intelligence becomes better at replacing traditional deep understanding of content."
www.psychologytoday.com/us/blog/the-...
Teaching a Kid to Think Like a Genius
Every child can develop Leonardo's genius-level thinking through cross-domain connections. Parents can teach these strategies to unlock this potential.
www.psychologytoday.com
July 10, 2025 at 10:28 PM
Cross-domain thinking - a key to innovation:
"The capacity to see connections across different domains of knowledge is extremely valuable for kids today, as artificial intelligence becomes better at replacing traditional deep understanding of content."
www.psychologytoday.com/us/blog/the-...
"The capacity to see connections across different domains of knowledge is extremely valuable for kids today, as artificial intelligence becomes better at replacing traditional deep understanding of content."
www.psychologytoday.com/us/blog/the-...
To maintain quality when using AI for software, a thorough review is required. Fortunately, you can ask the AI to explain itself while you review! Make sure you understand every line it writes. You might learn something new and simply asking often prompts the AI to identify and fix its own bugs!
July 8, 2025 at 10:12 PM
To maintain quality when using AI for software, a thorough review is required. Fortunately, you can ask the AI to explain itself while you review! Make sure you understand every line it writes. You might learn something new and simply asking often prompts the AI to identify and fix its own bugs!
If you're new to using LLM tools for software engineering, I highly recommend starting by leaning into Test Driven Development, where you write the tests, then ask the AI to implement the feature, referencing the tests for expected behavior. Cursor IDE models can run your tests to self-validate.
July 5, 2025 at 1:27 AM
If you're new to using LLM tools for software engineering, I highly recommend starting by leaning into Test Driven Development, where you write the tests, then ask the AI to implement the feature, referencing the tests for expected behavior. Cursor IDE models can run your tests to self-validate.
My team and I have been working with Cursor IDE @cursor.com.web.brid.gy for several months now. What it lacks in traditional IDE richness, it makes up for in quality AI integration, which is a net development accelerator. I can post some tips for getting the most out of it over the coming days.
July 2, 2025 at 3:49 AM
My team and I have been working with Cursor IDE @cursor.com.web.brid.gy for several months now. What it lacks in traditional IDE richness, it makes up for in quality AI integration, which is a net development accelerator. I can post some tips for getting the most out of it over the coming days.
Critical thought, too:
"To stay relevant, educators must move students beyond passive learning toward empathy, curiosity, adaptability and the kind of moral reasoning no algorithm can replicate. These are the skills that future employers and society will value most."
www.forbes.com/sites/kolawo...
"To stay relevant, educators must move students beyond passive learning toward empathy, curiosity, adaptability and the kind of moral reasoning no algorithm can replicate. These are the skills that future employers and society will value most."
www.forbes.com/sites/kolawo...
What AI Still Can’t Do — And Why It Matters More Than Ever
Machines can answer questions, but only people can ask the right ones. Here's why that matters more than ever in a world driven by AI.
www.forbes.com
June 28, 2025 at 6:12 PM
Critical thought, too:
"To stay relevant, educators must move students beyond passive learning toward empathy, curiosity, adaptability and the kind of moral reasoning no algorithm can replicate. These are the skills that future employers and society will value most."
www.forbes.com/sites/kolawo...
"To stay relevant, educators must move students beyond passive learning toward empathy, curiosity, adaptability and the kind of moral reasoning no algorithm can replicate. These are the skills that future employers and society will value most."
www.forbes.com/sites/kolawo...
Productivity lesson from 15 years ago, in @hollycummins.com recent convo on InfoQ:
"Okay, we're going to have a sprint where we are not going to be project managing this. You just do what you want to do. Some of our best features came from those time off sprints."
www.infoq.com/podcasts/pro...
"Okay, we're going to have a sprint where we are not going to be project managing this. You just do what you want to do. Some of our best features came from those time off sprints."
www.infoq.com/podcasts/pro...
Productivity Through Play: Why Messing Around Makes Better Software Engineers
In this podcast, Shane Hastie, Lead Editor for Culture & Methods, spoke to Holly Cummins about productivity in creative knowledge work like software engineering. She talks about how "messing around a...
www.infoq.com
June 27, 2025 at 3:17 AM
Productivity lesson from 15 years ago, in @hollycummins.com recent convo on InfoQ:
"Okay, we're going to have a sprint where we are not going to be project managing this. You just do what you want to do. Some of our best features came from those time off sprints."
www.infoq.com/podcasts/pro...
"Okay, we're going to have a sprint where we are not going to be project managing this. You just do what you want to do. Some of our best features came from those time off sprints."
www.infoq.com/podcasts/pro...