forward deployed engineer @wildchildai
Bangkok
Joined November 2022
- Tweets14.4K
- Following433
- Followers7.9K
- Likes15.1K
Pinned Tweet
No, I am not
ex-Meta
ex-Netflix
ex-Google
or any other ex's.
I was born in Iran, Tehran.
I had no friends because we had to travel every 3 months due to my dad's work.
My unfair advantage?
Sheer will!
I once told my uncle, "I'm going to make $6,000/month" one day, and he laughed at me.
He said, "It's good to have dreams, but they're just dreams."
Now I'm making between $12-$15k/month every month.
I bought a house and an SUV for my mom, helped my brother get into Webflow boot camp, and invested in real estate in multiple countries. I've launched more than 10 businesses and generated over $10m+ dollars for my clients over the past 10 years.
If I can do it, you can do it.
If you believe, you, like me, have 2 hands.
You can do it!
I think genuine human experiences are going to become more valuable as AI makes content easier to produce.
Everyone will be able to create beautiful videos, edit them, and build an atmosphere around a story that would've taken an entire production team just a few years ago.
But the actual experience behind that story is much harder to reproduce.
I suspect we'll see more people sharing very personal, sometimes niche experiences. Things that might seem ordinary to the person living them but are genuinely interesting to someone who has never experienced anything like them.
People want something they can relate to, or something that gives them a glimpse into a life they've never lived.
AI will make it much easier to tell those stories well. But having something real to tell might become the most important part.
馃捀 First @YouTube money!
Well second after a decade because I was a YouTuber from 2008-2013 with Panda Mix Show
But this is from my vlogging
About US$833
86
I think we're still very early in figuring out what the UI and UX of AI should look like.
Most of what we have today is still built around a chat interface. You type something, the model responds, and you go back and forth. It works, but I don't think this is where we'll end up.
Grokbot is getting pretty close to something interesting, but I think we can do much better.
Over the next 5 to 10 years, I think one of the biggest opportunities in consumer software will be figuring out how people actually want to interact with AI.
Not necessarily who has the best model, but who builds the best experience around it.
I suspect the companies that figure this out will win a large part of the next generation of consumer apps.
We're still so early.
Why is every AI a texting app?
We came up with an entirely new way to interact with AI. No switching apps, no texting, just a little barnacle that sits on your screen ready to help you out.
Waitlist opens today at barnacle.qa
1
94
Good intuition is just accumulated failure.
I actually like working on bad ideas.
The more bad ideas you work on, the sooner you get to the good ones.
We've seen this with so many indie hackers and founders. They spend years building questionable products, trying things that don't work, and chasing ideas that go nowhere.
But if they keep building, something interesting happens. They start developing better judgment. They learn what people actually want, what doesn't work, and which problems are worth solving.
A lot of what looks like good intuition is just accumulated failure.
I don't think those bad ideas were necessarily a waste of time. They were part of how you got to the good ones.
The important thing is to keep moving and not get too attached to any single idea.
82
I actually like working on bad ideas.
The more bad ideas you work on, the sooner you get to the good ones.
We've seen this with so many indie hackers and founders. They spend years building questionable products, trying things that don't work, and chasing ideas that go nowhere.
But if they keep building, something interesting happens. They start developing better judgment. They learn what people actually want, what doesn't work, and which problems are worth solving.
A lot of what looks like good intuition is just accumulated failure.
I don't think those bad ideas were necessarily a waste of time. They were part of how you got to the good ones.
The important thing is to keep moving and not get too attached to any single idea.
First successful test of session handoff across machines with OMP.
I started a session on Windows to inspect the top running processes, then moved that session over to my Linux machine. It worked.
One interesting detail: the original model wasn't available on the destination machine. OMP detected that it couldn't restore the model, showed a warning, and fell back to an available model.
This is exactly the kind of behavior I want. The session shouldn't be tied to a specific machine or even a specific model.
Still early, but it's a good first step toward moving active agent sessions between machines rather than just keeping them alive with something like tmux.
I've been working on something around OMP that I think is missing from how we use coding agents today: handing off active sessions between machines.
Right now, most tools solve this through session persistence. You start an agent on a machine, keep the process alive with something like tmux, and reconnect to it later. Rex has a similar approach.
But that's not quite what I want.
For example, I'm working on my laptop and need to close it. I want to hand the entire session over to my VPS or Mac Mini through Tailscale and have the agent continue working there.
Not reconnect to the laptop. Not keep the laptop awake. Actually move the session and continue execution on another machine.
There are some interesting problems here. The conversation history is only one part of the session. There's also the working directory, uncommitted changes, running processes, tool state, and the execution environment. Moving all of that reliably is a different problem from simply restoring a conversation.
I'm experimenting with this using OMP and Ghostty.
The goal is pretty simple: I should be able to start working on any machine, hand the session off to another, and pick up exactly where I left off without thinking about which machine is doing the work.
I think we're going to see more of this as coding agents become long-running processes rather than something we start and stop inside a terminal.
The machine should be an execution resource, not something the session is permanently attached to.
97
I've been working on something around OMP that I think is missing from how we use coding agents today: handing off active sessions between machines.
Right now, most tools solve this through session persistence. You start an agent on a machine, keep the process alive with something like tmux, and reconnect to it later. Rex has a similar approach.
But that's not quite what I want.
For example, I'm working on my laptop and need to close it. I want to hand the entire session over to my VPS or Mac Mini through Tailscale and have the agent continue working there.
Not reconnect to the laptop. Not keep the laptop awake. Actually move the session and continue execution on another machine.
There are some interesting problems here. The conversation history is only one part of the session. There's also the working directory, uncommitted changes, running processes, tool state, and the execution environment. Moving all of that reliably is a different problem from simply restoring a conversation.
I'm experimenting with this using OMP and Ghostty.
The goal is pretty simple: I should be able to start working on any machine, hand the session off to another, and pick up exactly where I left off without thinking about which machine is doing the work.
I think we're going to see more of this as coding agents become long-running processes rather than something we start and stop inside a terminal.
The machine should be an execution resource, not something the session is permanently attached to.
If you're building on virtual machines / Mac minis or over tail scale go with:
T3 Code
If you're building on terminal (Rex/Ghostty) go with:
Omp, what I like about this is that it load balances your subscriptions very easily.
With T3 Code however you need sub router plugin to switch between your Codex/Claude Code subscriptions.
3
234
Having so much compute and AI at our hands, we are still seeing a lack of quality products and taste. I was playing with Ghostty and Rex and I could barely find anything similar to them in any other platform.
1
212
"How to Win Friends and Influence People" is one of the best book that I ever read when I was like 15.
205
Babbage is finally here! A Debian 13 desktop built with Nix, i3 tiling, one flat menu bar, every coding agent preinstalled (Codex, Grok Bot, T3 Code, DeepSeek Harness, Tern), offline voice typing, one-command installs, and a hundred other details tuned to the pixel.
5
281
Many enterprises won't use Claude or other third-party AI services directly due to compliance and data governance requirements.
The bigger opportunity is to go niche, build internal AI tools around their workflows, and deploy them in their own cloud infrastructure.
If you're in a broad category it's time to go niche.
Claude may be eating Legal, Design and Finance, but it will take them a while before they get to "the AI that helps you choose the right vegan-friendly fishing tackle for Hudson County, New Jersey".
Moat.
1
332
