Simulation environments for verifying mission critical software. Backed by @ycombinator and @solofounders
Simulation
Joined June 2025
- Tweets63
- Following4
- Followers1K
- Likes215
Pinned Tweet
Introducing Workers IO
Agents are making it possible to turn ideas into software faster than ever. To keep up with that, we need better ways to understand how our software will behave in real-world.
Workers IO runs your entire system through millions of simulated futures, exploring what happens when machines fail, networks slow down, and events arrive in the wrong order. It searches for the combinations that break your system before your customers encounter them.
Every failure it encounters is perfectly reproducible, helping teams understand what happened, work on a fix, and verify against the same scenario.
This ability changes what teams are willing to take on. Difficult engineering decisions become things that you can explore and verify.
That’s the future we’re working toward: small teams building increasingly ambitious systems, with the evidence to stand behind.
I can’t wait to see what you build with that confidence.
Workers IO retweeted
Internet really is falling apart, a year ago my @solofounders cohort and I first spoke about "fixing the internet"
The cracks are finally starting to show in a big way. Until now, teams could get away with sloppy engineering and poor security practices purely because the cost of compromising systems was super high.
That is no longer the case, now all it takes is a few tokens a single gap in your system for it to fall apart.
eMudra incident is not going to be the last one and it'll be irresponsible for teams to not have dedicated budget just for security and correctness of their systems.
@workersio will have a big part to play in this!
Workers IO retweeted
So many teams have traded understanding and safety for velocity.
They didn't have to, you really can have it all!
Workers IO retweeted
Anyone I know looking to try out a new DST product? they're small but have already closed a few interesting logos, seem to have a good product, and very very affordable
What does it take to verify mission-critical software?
One of the most difficult things when it comes to testing complex systems is figuring out all the ways it can fail.
How your system can behave in production is not just limited to the code, you also need to look at the behavior of the infrastructure that it runs on and how your customers are using it.
At Workers IO, we’ve put a lot of work into a workload generator that turns this mental exercise into a compute problem.
Here’s how it works.
How to test for failures that only happen in production?
Most teams test their system in a perfect world. However, in production, packets get lost, machines crash and even disks fail.
To verify against these failures, you need a fault model: every part your system depends on, from the network, system calls, memory, disk, and the ways each one can fail. Writing one takes deep knowledge of every layer, and most teams never get to it.
At Workers IO, we build that fault model for you, then run your entire system through a different mix of failures every time. When things breaks, you get the exact scenario that caused it with perfect reproducibility.
How to test for failures that only happen in production?
Most teams test their system in a perfect world. However, in production, packets get lost, machines crash and even disks fail.
To verify against these failures, you need a fault model: every part your system depends on, from the network, system calls, memory, disk, and the ways each one can fail. Writing one takes deep knowledge of every layer, and most teams never get to it.
At Workers IO, we build that fault model for you, then run your entire system through a different mix of failures every time. When things breaks, you get the exact scenario that caused it with perfect reproducibility.
Workers IO retweeted
Our workload generator just found a really interesting failure. It simulates people typing into many remote sessions at once, over a network it deliberately makes slow, lossy or out of order.
The system prioritizes what it sends, so the echo of a keystroke goes out ahead of a flood of logs. But when the network got slower after it connected, we kept sending at the old speed. The extra data piled up in the network, so every keystroke waited behind it and lagged by about half a second.
Most teams never write scenarios like this, let alone test against them. It's just a default for us :)
What does it take to verify mission-critical software?
One of the most difficult things when it comes to testing complex systems is figuring out all the ways it can fail.
How your system can behave in production is not just limited to the code, you also need to look at the behavior of the infrastructure that it runs on and how your customers are using it.
At Workers IO, we’ve put a lot of work into a workload generator that turns this mental exercise into a compute problem.
Here’s how it works.
Workers IO retweeted
formal verification is one of the underdeveloped fields in the modern tech, will probably be huge unlock for AI in mission critical software
What does it take to verify mission-critical software?
One of the most difficult things when it comes to testing complex systems is figuring out all the ways it can fail.
How your system can behave in production is not just limited to the code, you also need to look at the behavior of the infrastructure that it runs on and how your customers are using it.
At Workers IO, we’ve put a lot of work into a workload generator that turns this mental exercise into a compute problem.
Here’s how it works.
What does it take to verify mission-critical software?
One of the most difficult things when it comes to testing complex systems is figuring out all the ways it can fail.
How your system can behave in production is not just limited to the code, you also need to look at the behavior of the infrastructure that it runs on and how your customers are using it.
At Workers IO, we’ve put a lot of work into a workload generator that turns this mental exercise into a compute problem.
Here’s how it works.
Workers IO retweeted
One of the hardest things we do at @workersio is assigning a severity to a finding.
It’s a big responsibility because we must to do a better job than our customers.
If we downplay a finding it may end up getting deprioritized and lead to a more serious incident later on, but it’s also just as easy to annoy people if we overstate things.
The biggest advantage we have is, every finding is used to chain things and explore new behavior, makes this job a lot easier than looking at things in isolation.
Workers IO retweeted
Crazy this video got us to 955 followers in a day.
Thanks @claudeai for helping out a solo founder ;)
Introducing Workers IO
Agents are making it possible to turn ideas into software faster than ever. To keep up with that, we need better ways to understand how our software will behave in real-world.
Workers IO runs your entire system through millions of simulated futures, exploring what happens when machines fail, networks slow down, and events arrive in the wrong order. It searches for the combinations that break your system before your customers encounter them.
Every failure it encounters is perfectly reproducible, helping teams understand what happened, work on a fix, and verify against the same scenario.
This ability changes what teams are willing to take on. Difficult engineering decisions become things that you can explore and verify.
That’s the future we’re working toward: small teams building increasingly ambitious systems, with the evidence to stand behind.
I can’t wait to see what you build with that confidence.
Workers IO retweeted
DST coming soon to Durable Actors! Shout out to @chaaai and @workersio
Introducing Workers IO
Agents are making it possible to turn ideas into software faster than ever. To keep up with that, we need better ways to understand how our software will behave in real-world.
Workers IO runs your entire system through millions of simulated futures, exploring what happens when machines fail, networks slow down, and events arrive in the wrong order. It searches for the combinations that break your system before your customers encounter them.
Every failure it encounters is perfectly reproducible, helping teams understand what happened, work on a fix, and verify against the same scenario.
This ability changes what teams are willing to take on. Difficult engineering decisions become things that you can explore and verify.
That’s the future we’re working toward: small teams building increasingly ambitious systems, with the evidence to stand behind.
I can’t wait to see what you build with that confidence.
Workers IO retweeted
Rick won us another contract!
Customer: "I never really understood what you do till now, can you onboard us right away?"
What we are building will change how mission-critical software is built and verified
Here's Rick & Morty to tell you all about @workersio
Workers IO retweeted
cartoon explainers are the new meta
How did he even make this
What we are building will change how mission-critical software is built and verified
Here's Rick & Morty to tell you all about @workersio
Who better to explain deterministic simulation testing than Rick himself
What we are building will change how mission-critical software is built and verified
Here's Rick & Morty to tell you all about @workersio
Workers IO retweeted
Tokens alone are not going to cut it, software verification is very much a research problem just like AI safety.
We need better tools, there is a reason why Rust is overtaking Python in a lot of cases.
Agents make choices too and they will pick the best tools and techniques to build verifiably correct software.
Security will become a larger and larger function in software companies. Security is verification engineering (eg: “my code is probably memory-safe”), as well as capital allocation (“what surface should I throw most tokens at?”)
For startups, it’s a challenge and opportunity. As the world’s paranoia grows given ever-more-sophisticated AI adversaries, how can you trust a 2-person-and-a-dog company?
But conversely, given the dire state of global cybersecurity and increasing reliance on centralization, as always, there’re vast areas that small teams can now disrupt that I’m excited to invest in.
Workers IO retweeted
The equivalent of this in software engineering is not another agent reviewing the output and telling you it's good.
It's precise specifications, proofs, model checking and simulating real production scenarios as proof of work.
There is no way any of us are going to look at code the same way again.
We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks:
Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better:
Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better:
Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better:
Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work!
In summary:
- As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding.
- Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.
