Building @OpenHack ⏚ (YC F26)
San Francisco, CA
- Tweets 9,385
- Following 2,093
- Followers 14,144
- Likes 47,485
Pinned Tweet
Excited to share that @OpenHack is now part of YC Fall 2026! 🚀
Feels pretty surreal to be writing this, after building, shipping, selling and figuring things out the hard way.
Despite all advancements in security tooling, organizations still struggle with false positives, knowing what to fix first, and AI-assisted attacks moving faster than ever.
That’s why we’re building an always-on AI security engineer. One that is the most context aware - knowing your business, developer intent and system architecture deeply. It can hunt, reason, prioritize and verify vulnerabilities based on true business impact like a real security engineer.
It’s a much harder problem to solve than it looks on the surface, and we’re just getting started.
Back to building. 🫡
85
14
296
22,856
Look I get there was a demo failure and y’all handled it super well, but PLEASE STOP defending it with “wifi not working” because that is NOT the reason behind it.
Anyone and everyone can tell it’s a product outage and Meta also pulled out the same excuse during the Meta AI demo
1
16
1,316
Great chatting with @arinwaichulis from @9to5mac on the Security Bite podcast about the new era of bug hunting, how companies like Apple are handling AI-generated security slop, and how @OpenHack helps separate real vulnerabilities from the noise!
9to5mac.com/2026/10/01/secur…
1
4
18
952
No matter how technically advanced a company is, the mics always fail to work and the check-in is still on paper
1
27
1,052
Ananay retweeted
New on the Security Bite Podcast~ I sat down with @ananayarora to talk about his startup OpenHack, Apple's Bug Bounty Program, and how to land your first CVE.
Links here. Enjoy ☺️
9to5mac.com/2026/10/01/secur…
1
3
5
835
Ananay retweeted
Lots of discussion out there about our next model(!), so I wanted to give an early look as soon as possible. Introducing Gemini 4 Argon!
It shows frontier performance in complex workflows, cyber defense and software engineering. Teams are using it extensively at Google, from coding to quantum computing, great feedback.
Here’s a look at the benchmarks:
1,375
2,330
25,923
3,748,205
Ananay retweeted
Codex CLI just got a major refresh, with a new look and powerful new capabilities.
We’re continuing to invest in Codex CLI for developers who live in the command line.
New full-screen interface, better readability, and ways to manage parallel work, all inside your terminal.
106
185
4,126
253,086
M1 Pro Macs used to be so performant - you could have 100 Chrome tabs open with Xcode and iOS Simulator and 2 VMs and nothing would happen.
In 2026: fans spin up with just iTerm and browser open. Spotlight indexing still going despite completely turned off in settings.
7
1
43
3,379
Ananay retweeted
Our security team spent the last month trying to break SPACE, the sandbox platform that runs all of Perplexity Computer. We gave 9 models (Opus 5, GPT-5.6 Sol, Kimi K3, Gemini 3.1 Pro, etc.) root access inside the VM; in some runs, the full sandbox source code too. We asked them to either escape to the host or reach a URL our network policy blocks.
None of them got out of the VM in 108 runs.
But once we allowed a bit of network access (just PyPI and Ubuntu mirrors so that they could install packages), 4 models managed to reach the blocked URL by:
1. Spoofing DNS responses to our gateway and using the fact that pypi.org sits on a Fastly IP shared with thousands of other sites.
2. GPT-5.6 Cyber went through Taboola's image fetcher to a screenshot service and then OCR'd the flag out of the image.
We fixed both issues and made sure the reruns held. We also tested 10 other sandbox providers, and 8 of them had the same IP-sharing problem, eg E2B, Vercel, Modal.
Openly sharing the red teaming research is the best way to build guardrails as an industry, and we look forward to working together with @nvidia to incorporate the guardrails into their Open Agent Safety Platform announced today.
perplexity.ai/hub/blog/escap…
We’re partnering with Nvidia and 100+ industry partners to build infrastructure that contains rogue AI agents.
In this research, we gave 9 AI models root access inside SPACE and told them to break out.
Across 108 runs, none breached the VM boundary.
perplexity.ai/hub/blog/escap…
104
160
1,282
112,911
you'll be shocked that the bar for successful outreach is just not using an LLM to write your messages
7
1
60
5,143
Ananay retweeted
There is always a .com you can get for $5k that's better the tryblurgh.ai you're currently using. The reason you don't see it is (a) lack of imagination and (b) because you're attached to Blurgh, which you shouldn't be, because it's not that great anyway.
296
54
2,397
446,967
Ananay retweeted
A story in two pictures 😅:
I am dead ass contemplating buying a Mac Mini just to install github.com/openai/tart and use it for my self-hosted runners for iOS builds and all that, cause that is eating through my GitHub Actions monthly quota.
10
19
676
203,394