Samir Sangani

Dad trying to learn and explore how AI reshapes real-world systems. Building fun apps with daughter :) 🇨🇦

[Samir Sangani @SSangani]

@ClaudeDevs @ClaudeDevs Please give me a way to set when my limit resets since the reset you just did will not help me in any way and this happens everytime 🥲

[Samir Sangani @SSangani]

@theo Yes, and it always resets when I have a few hours left in the actual reset 🥲

[GREG ISENBERG @gregisenberg]

When I first became a dad I was genuinely worried my career would suffer.

The opposite happened. 3 things changed that I wasn't expecting.

First, a child cuts the filler from your life instantly.

I used to sit at my desk for 14 hours and feel like I was crushing it when in reality maybe 4 of those hours were actual work and the rest was meetings that didn't need to happen, scroll sessions I told myself were research, and "quick calls" that turned into 90 minutes of nothing. A child deletes all of that overnight.

Because you literally don't have the time anymore. Every hour matters in a way it didn't before. You could be with your kid, working on your startup, exercising, having dinner with your wife, sleeping. When your time is actually full of things you care about, the filler can't survive. I'm shipping more now than before my kid was born. Half the meetings. Faster decisions.

I stopped saying yes to things out of politeness because my time has a very real cost now that I can feel in my bones.

Second, your risk tolerance goes up, not down.

Everyone assumes having a kid makes you play it safe. For me it created this urgency to build something real while my kid is young enough to not remember the hard parts. That urgency is more useful than any productivity system I've ever tried.

Third, your thinking just gets clearer.

I don't know how else to explain it. You stop deliberating for days and just make the call. You stop chasing every opportunity and only chase the ones that actually excite you.

Something about being responsible for another human being gives you this filter that cuts through the noise instantly. Before my kid, I'd go back and forth on a decision for a week. Now I make it by lunch and move on.

I used to think having a kid was the thing I'd do after I built the company. Turns out the kid made me better at building the company. Wish someone had told me that sooner. So I'm telling you.

I know this sounds like something a new dad says to justify it. I thought the same thing when other dads told me. Then it happened to me and I understood.

I think you will too.

[Lee Robinson @leerob]

"Engineering, product, and design are all merging into a 'builder' role"

Yeah... I'm not so sure. This feels like an oversimplification and podcast talking point. Reality is a lot more complex.

Even with 1000 "Member of Technical Staff" titles, someone still has to wake up and care 100x more about Product or Design than anyone else. It is their Main Thing™

That's not to say MTS titles are universally bad, but I think they're an example of this 'builder' talking point that's become bastardized.

AI and coding agents have made generating code easy and yet... you're in for a world of pain if non-engineers ship a bunch of slop and don't have great engineers to tame the complexity.

The SF hivemind has a tendency to overfit what works at startups for every company. And to be fair, sometimes this is true! Startups can be a leading indicator for how the industry is changing and often cause disruption.

However, it is going to be incredibly hard to disrupt the extremely human parts of corporate jobs. You really think there's going to be a PM who also does some engineering and design on the side at JPMorgan Chase?

This is true for the simple parts of most jobs, like people wanting to have ownership over something and do good work, move up a career ladder, support their family, get paid well, make an honest living...

And also the hard parts: internal politics, some critical business system that has a bus factor of 1 which has been running for 15 years and isn't documented anywhere because it's that guy's job security. The real world has a lot of this stuff.

It's easy to pontificate about all roles collapsing but it's actually really nice to have a specific person or team who is an expert in one thing that you can work with. I don't expect that to change. Further, I think AI disruption to knowledge work will take decades to play out because it is more fundamental to the human condition (e.g. sociological/organizational) than pure intelligence.

[Samir Sangani @SSangani]

I recently saw on my Profile that I am following more than 1200 creators, and it really surprised me, so I asked my daughter whom I sometimes give my phone so that she can catch up on latest trends and keep up to date on the craziness of AI. She told me it was her and she said isn't that good, the more people we follow the more our knowledge will increase (she said what's the point of followers 😅). This really made me stop for a second and reflect on how we are rushing to reach posting milestones to get paid vs a pure mind that is hungry for knowledge and does not care about anything else. I just wanted to thank all the amazing female creators on @X who keep inspiring her 🙏

[Samir Sangani @SSangani]

I really love the Today's News feature on @X, it really helps users like me who sometimes do not have the time to go through the posts get a quick overview of the trending posts and catch up with the updates

[Samir Sangani @SSangani]

I usually do not say anything bad about any models since for me I always find a way to work around most of the issues. But @ClaudeDevs Claude Code has become so slow that it has become frustrating now. As you can see in the image, it has been working since more than 5 mins and has generated less than 500 tokens and the token count has not moved since the past 4 mins, it seems that I am in some kind of queue waiting for my turn after which it will start working again?

[Samir Sangani @SSangani]

@gregisenberg Have created an automation system that uses the API to fetch and analyze them and finally put the analysis in Obsidian, but now my Obsidian has become difficult to consume :) Hopefully it builds enough knowledge that I can probe one day as needed.

[Samir Sangani @SSangani]

I have been seeing this weird thing in @cursor_ai wherein all skills are being loaded in the context. I sent a single prompt (Composer 2.5) and my context was filled up by skills. Is this correct, I though that only the name and description of the skill was sent and the actual skills are loaded as needed? Is this not the case? I tried the same prompt in Codex and the context usage was drastically less.

[Samir Sangani @SSangani]

One thing AI products still get wrong: Mid-session conversation compaction without clear user control.

If context is about to be compressed, let me choose when it happens (even if the context limit is still a bit further from when compaction happens) or preview the impact before I send the next prompt. Otherwise the session suddenly feels unstable and opaque. I can see the context now in Claude Code clearly but cannot see it anymore in Codex app. However, in both cases, its always a gamble on how much will the prompt I am giving consume tokens and a hope that there is no compaction that occurs (usually worse when it happens just before the task is about to finish).

[Samir Sangani @SSangani]

Has anyone compared the $200 plan for Cursor vs ChatGPT vs Claude? I have seen a few comparisons between ChatGPT and Claude but they seem to be just users complaining on usage limits. How are the three different from the perspective of developing a production app - which one would you trust more and why? I understand that the harness is different and would thus provide different usage patterns, however which one gives you more confidence in using it for every day use?

[Samir Sangani @SSangani]

https://t.co/lVKt9QOOs2

[Samir Sangani @SSangani]

@bcherny I have been trying to use Claude Code to address the npm vulnerabilities found almost everyday now. One of the challenges I have been having doing that is the agent gets blocked by the content filter (especially if I provide it an article or X post mentioning the information). Has anyone found a way around this?

Are there other methods one can use to keep up to date with these security vulnerabilities using Claude Code or Codex or Cursor, any way to automate it?

[Samir Sangani @SSangani]

Have been trying both Grok Imagine and Google Omni with my kids and it has been a lot of fun bringing their images to life, I would suggest everyone to try it at least once. My kids seem to prefer Grok @imagine compared to Google Omni since it keeps the original sketches intact. Omni completely changed the image, but we loved how it burned the paper and brought it to life. Omni just released (with really bad usage limits 🙁) so we hope it will get better with time. Enjoy the generations below (Imagine on left and Omni on right).

[Samir Sangani @SSangani]

@elonmusk @elonmusk Please bring back the deal so that we can try it. Missed it due to long weekend.

[Samir Sangani @SSangani]

https://t.co/u62Dj7GOm2