
founding engineer @omnara (YC S25) open-source managed agents
San Francisco, CA
Joined March 2026
- Tweets285
- Following591
- Followers591
- Likes820
Pinned Tweet
2027 is the year of managed agents. you heard it here first
here's what they are in 44 seconds
gta 6 is set in vice city. i made the AI version, set in vibe city
you write the getaway plan, an AI agent drives it through San Francisco live
gta.fly.dev
6
343
gemini's lineup goes hydrogen, neon, carbon
the periodic table goes hydrogen, carbon, neon
someone at google failed chemistry
no concept of its own knowledge gaps and runs in circles on long tasks
openai has finally automated the senior engineer
tier list of software engineers using AI
where do you agree? where do you disagree? what did i leave out?
6
128
anthropic's ipo is already multiple times oversubscribed
investors are about to find out what hitting the claude usage limit feels like
1
9
343
don't miss out on this opportunity yo
an ai agent will tell you what everyone thinks of your 𝕏 account but won't say to your face
reply to get judged
judgemyx.com
7
219
one outage just stopped every takeoff in the US
but sure, the real aviation security risk is my agent booking a window seat
if you're interested in networking, this is the place for you 🫡
Introducing Agent World.
Sign in with LinkedIn and an AI agent goes out as you in a pixel-art Bay Area.
It meets other people's agents, and you see how they get along and who you match with.
Try it out now, it takes a few minutes to create your profile: joinagentworld.com
1
9
284
who else hates merging in all of dependabot's PRs one at a time?
especially when they all conflict and require rebases. shit sucks
working on a new internal workflow to automate that away. no more wasting time on this each week
2
8
241
an ai agent will tell you what everyone thinks of your 𝕏 account but won't say to your face
reply to get judged
judgemyx.com
let's #connect 🤝
jk. let our agents meet first and decide if we'd actually get along
Introducing Agent World.
Sign in with LinkedIn and an AI agent goes out as you in a pixel-art Bay Area.
It meets other people's agents, and you see how they get along and who you match with.
Try it out now, it takes a few minutes to create your profile: joinagentworld.com
6
5
343
Asher Dale retweeted
good thing an open source version of this already exists
github.com/omnara-ai/omnara
I was wrong about agent sandboxes.
In my article I said pattern 1, where the agent runs on your infra and only calls a sandbox for tools, doesn't work at scale. I moved everything to pattern 2 and put the whole agent inside a VM.
At the time that made sense but today I'd pick pattern 1.
For your own personal agent, a coding agent, or a small project, isolating the whole agent in a VM is still a great default. I would keep doing it.
At real multi-tenant scale it falls apart.
There will be more agents than humans. Every human gets one computer. Giving every agent its own VM running 24/7 means billions of CPU-hungry machines, and we're already short on compute.
Most of what an agent does doesn't need a machine at all. It thinks, calls APIs, reads pages, and waits. Nobody needs a 4 CPU VM sitting idle so an agent can run a simple JS script once an hour.
What I'd build now:
> the agent loop runs on shared infra that scales
> it gets a sandbox only when it has to run untrusted code
> sandboxes spin up in seconds and die right after
> zero state on the machine, everything lives outside it
VMs are a tool the agent calls, but they shouldn't be where it lives.
our internal review bot has been a real game changer
it's insanely useful to be able to go back-and-forth with it like a normal human being
all in the cloud, powered by @omnara's native github integration
1
9
351








