kaushik saurabh

Android Engineer/Ex-Doordash/Ex-Snapchat/Ex-Tinder

[kaushik saurabh @roideuniverse]

@flipper_net I placed an order, and it been more than 10 days and it has not even shipped.

[kaushik saurabh @roideuniverse]

@flipper_net It's working now

[kaushik saurabh @roideuniverse]

@flipper_net

[kaushik saurabh @roideuniverse]

@flipper_net Is it down?

[kaushik saurabh @roideuniverse]

@jescalan Lol this was funny 🤣

[Brian Attwell @attwellbrian]

Earlier this quarter, I bet one of our EMs he couldn't beat the AI productivity gains seen in credible field studies (ie, Demirer et al). The odds were against him.

I lost 🤦

We sat down at Atoms HQ to unpack what worked, what surprised us, and where to go next:

[Ahmad @TheAhmadOsman]

> be anthropic engineer
> realize long-running agents still have goldfish memory
> every new context window = new intern who forgot everything from yesterday
> project goes from “build a clone of ChatGPT” to “why is half the frontend missing again?”

> agents try to one-shot entire apps
> run out of context mid-feature
> next session wakes up like “boss… who touched the router folder? why is the server on fire?”

> other times claude walks in
> sees 3 buttons rendered
> declares the whole project complete
> packs up its laptop
> goes home

> humans don’t work like this
> engineers leave breadcrumbs
> notes, git commits, tests, todo lists
> “here’s what I did, here’s what’s next, don’t break the login page again please”

> so anthropic builds a harness based on that
> two-agent setup: initializer agent, coding agent

> initializer agent = the senior dev on day one
> sets up:
> – _init.sh
> – claude-progress.txt
> – feature_list.json (200+ features, all marked failing)
> – the first git commit
> basically: “here’s the blueprint, don’t get cute”

> coding agent = the worker bee
> every session:
> – read the progress
> – read the git log
> – read the feature list
> – pick ONE feature
> – implement it
> – test it end-to-end as an actual user
> – commit code
> – leave notes
> – do NOT break anything, or revert yourself

> incremental progress > chaos
> and forcing the agent to act like a real engineer = night and day difference

> testing was the big “aha”
> claude kept marking features done that absolutely were not done
> (“unit tests pass” ≠ “the app works”)

> give it browser automation + puppeteer
> claude suddenly starts catching bugs it introduced 5 minutes ago
> screenshots, clicks, actual user flows
> end-to-end or bust

> limitations still there
> puppeteer can’t show alert modals
> claude can’t see everything
> vision quirks remain
> but it’s way closer to real QA than “lol curl localhost:3000”

> typical session now looks like:
> “pwd”
> read progress
> read features
> read git log
> start server
> sanity test
> fix broken stuff
> choose next feature
> implement
> test
> commit
> leave breadcrumbs
> repeat

> four classic failure modes solved with structure:
> – agent declaring victory too early → feature list
> – messy environment → git + progress logs
> – premature ‘passes’ → real testing
> – agent forgot how to run app → _init.sh

> does it solve everything? no
> still open questions:
> single agent vs multi-agent division of labor
> maybe future = dedicated QA agent, cleanup agent, test writer agent
> maybe research workflows get similar scaffolding
> maybe finance models get their own version

> but the core insight stands:
> long-running agents don’t fail because they’re dumb
> they fail because we throw them into multi-session hell
> without giving them the engineering rituals humans rely on

> give them structure, tools, tests, logs, diffs
> they stop acting like goldfish
> and start acting like teammates

[Shervin @shervin]

Recall. Impeach. Prosecute.

California is burning—literally and figuratively. Decades of liberal mismanagement, woke incompetence, and defunding police and fire services have broken governance. How can there not be enough water for firefighters in a state with this much wealth?

This failure of leadership is staggering. Los Angeles Mayor Karen Bass is in Ghana while Angelenos face devastating loss. She should be impeached. The fires in the Palisades and Malibu are raging uncontrollably, and the federal government must act. President Biden, declare an emergency and deploy Navy firefighters and the Army Corps of Engineers now. If Trump were president, decisive action would already be underway.

I’ve had to step in personally with private firefighters and security people to help save friends’ homes—because the system has failed. Citizens shouldn’t have to do what their tax dollars are supposed to cover.

To the brave firefighters and first responders risking their lives: you deserve better. California deserves better.

This state is the Amalfi Coast of America—a jewel of breathtaking beauty. It’s time to turn the tide.

Make California Great Again.

@POTUS @realDonaldTrump

[kaushik saurabh @roideuniverse]

I was there with my wife last to last year and she was like forget living here we are not coming back here again.

[kaushik saurabh @roideuniverse]

This is the current state of San Francisco which used to be one of the best cities but now is like Gotham City.
It has been like years now.
I just do not understand how and why it has gone so downhill.

https://t.co/Ylt6KwztdK

[kaushik saurabh @roideuniverse]

@baavri @airindia This isn't enough. Sue @airindia
This is exploitation and corruption.

[Janet A. Carr @janetacarr]

Throughout the years I have hired a lot of developers

I have rarely met talented developers that didn't have side-projects of some kind.

if side projects are a problem for your company, you have a leadership problem at your company.

[kaushik saurabh @roideuniverse]

@Chewy @Chewy again 🤦‍♂️

[kaushik saurabh @roideuniverse]

Not sure if the Chewy delivery services are crazy or just mindless. First they leave the heavy box in the lobby and on top of that above the mailbox box. Happens everytime! Lol @Chewy

[Chris Banes @chrisbanes]

TIL that IntelliJ/Android Studio has a widescreen monitor setting. ↔

It lays out bottom docked panels to be displayed within any panels on the left/right. Basically, the bottom docked panels aren't displayed all the way on the edges now. I much prefer it.

[kaushik saurabh @roideuniverse]

@awilkinson @2percentJazz2 Very Cool

[DeL2000 @DeL2000]

This man is a HERO‼️ Francis Zuber ‘s headcam shows him rescuing a snowboarder who was buried head down in deep powder at Mt. Baker, Washington! Luckily, he spotted him as he almost passed him. #skiing #snowboarding

[B Graham Disciple @bgrahamdisciple]

My son, who is a software engineer, just told me that people who know software just had their Elon Musk realization.

[kaushik saurabh @roideuniverse]

@davidlubar I have had similar conversations with both Fedex and chewy and at the end had to give up on it

[kaushik saurabh @roideuniverse]

@AmazonHelp Thanks, it was delivered

[kaushik saurabh @roideuniverse]

When @amazon does not want to deliver on the weekend @AmazonHelp