Abejith

A lot of the emails I get from founders are now written in a hard-hitting journalistic style. I know they're written by AI, because no founder ever wrote this way before. And once you realize something is written by AI, it's hard not to ignore it.

Using SF Muni's sounds with @conductor_build is so pleasant. Same feeling when your train arrives as when your agent finishes something 😂

The best part of computer use with Codex is that you can continue working in parallel, and it doesn’t interrupt your flow. Basically, more versions of you are working behind, in the same machine.

Interesting.

HERMES AGENT IS ONE OF THE FIRST AI PROJECTS THAT ACTUALLY REMEMBERS EVERYTHING ACROSS SESSIONS AND GETS BETTER THE MORE YOU USE IT. Multi-layer memory, self-evolving skills, autonomous 24/7 agents and cross-session recall. It feels less like a tool and more like an operator.

thanks to this recent exciting update: x.com/grok/status/20…

When you plug @grok into hermes and X is now agentic.

I’ve been using Apple Music for over a decade now (surprisingly started using it on Android since its launch in 2015 on a student plan). In the early days, I’d see exactly the feature request that I put into the survey, going live within a week. It was a surreal product experience all along, and I still think it’s the best music experience. But this particular moment with AI coming into the mix can’t be described in words. Most of the feedback that I used to share is now possible to be done DIY in a few prompts, and I have an explosion of ideas of what’s possible. The experience curve has taken a sharp turn upwards, and it is one of the many areas where you see AI is not just a passing trend.

This is so exciting. I think I have a codex fever now.

Today we’re releasing Appshots in Codex. I love this feature! It makes it so easy to integrate Codex into everything I do on my Mac.

@steadwing @OpenAIDevs

5M tokens later Codex has a pet for @steadwing. Very impressed!

Karl the Fog rolling into Union Square

Here's a number that doesn't show up in most engineering dashboards. Each after-hours incident costs roughly 4–6 hours of productive engineering time the following day. Not the incident itself. The day after. 2 hours on post-mortem docs and Slack async. 1 hour in a meeting too tired to contribute. 1 hour of feature work that just doesn't happen. For an on-call engineer handling 3 incidents a week which nearly half of SREs report that's 50+ hours of lost productivity a month. This never appears in your MTTR dashboard. But it shows up in velocity and attrition. The real ROI of reducing on-call burden isn't faster incidents. It's preserving the other 99% of your team's engineering time.

Exciting times ahead! That was a good one on Codex at @OpenAI @AIBoomi

@p_naix 🙌

@Abejith Ideally should work. Just added, will let you know tomorrow if it works

GPay India supports Mercury Business Card. Crazy!!

I’ll be judging the AI Hackathon on May 23 in SF with @scalekitinc , @apify , and @EntireHQ . The prompt is simple: Take a workflow you still do manually, then automate it end-to-end. Pull the data. Enrich it. Take the next action. All with real user permissions, not a service account. Excited to see what gets built in a day. If you’re working on agents or workflow automation, come by. Registration link in replies!