Deepx (@Deepnsec) header
Deepx (@Deepnsec) profile picture

founder @ https://teapaw.org/t.co/pIG1tlL4ZP | building intelligence os for architects

London
Joined September 2019
astra said no, wtf!
1
23
"zero data retention" on LLM routers is a checkbox, not a guarantee @OpenRouter just emailed me: my prompts to minimax may have been read by a third party a researcher seeded canary urls in prompts. they got opened 40-130 hours later. even via novita, atlascloud and streamlake, which were quietly proxying to minimax ZDR is self-reported. nobody verifies it if your agents put api keys, customer data or repo urls in prompts, assume some of it was read what to do: - pin providers, never let the router pick - go direct to the model lab when you can - seed your own canary tokens - rotate anything sensitive
5
4
159
Replying to @Lantos1618
so true lol
1
33
my agent did my whole job hunt on @treg_ai while i got my iced latte. 5 roles, £120k median, the hiring manager. total bill: under 4 cents
🤝 Paid partnership
Introducing Jev @typesafeai powered jobs-api Try it at treg.to/jobs-api - One API to any jobs data - $0.03 to run a full job hunt - Real salary & Hiring manager data - Plugin to any agent for your dream job Fully open source, 0% markup Git Repo below 👇
2
4
10
344
respect for redoing the math in public after vercel's ceo called it out ran a small agent task on @Agent37HQ: whole run $0.028, sandbox part $0.0003. the llm was basically the entire bill lol
🤝 Paid partnership
The CEO of Vercel called us out. We claimed Agent37.com was 25x cheaper, and he said we didn't consider active CPU time. So we fixed the math.. We gave 100 coding agents a sandbox each and had them vibe code an app. Here's the cost per hour at just 5% active CPU: Vercel: $9.75 (14x) E2B: $16.56 (24x) Daytona: $16.56 (24x) Agent37: $0.68 We're starting a race to the bottom... see you there. Find a cheaper sandbox and we'll refund the difference.
9
384
Rockstar took 12 years for GTA6. Fable 5.2 took one Prompt!
Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges. They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.
2
2
8
488
the side project i almost didn't ship just made me money, zero ad spend btw, built with @t3dotcodes
2
1
6
140
stop cold emailing the biggest firm in your niche find whoever sits on the industry panel and call them first. my agent found one on a planning design panel today
🤝 Paid partnership
Introducing people search for Fable - find leads for cents - get and verify their work emails - pay per use across Apollo, Clay, Firecrawl and more composio.dev/instant/people-… start with up to 70 verified leads for free
2
3
8
143
no way @FishAudio built me a london site briefing agent for architects in 5.7 seconds from one comment 😭🏗️
🤝 Paid partnership
Introducing @FishAudio Live 3: a full-duplex voice model that listens, says "mm-hmm," and knows when to shut up. Enterprise? Book a demo for your free custom agent: s.fish.audio/vxhgs6 Reply with your use case - we'll build it free for the first 1,000 people.
3
6
140
Is Huang Xiaole (黄小乐) aka Alice Huang a 燕子 working for the CCP?
1
5
407
was expecting 6.2 Astra ;)
Introducing GPT‑6 Cyber: a major advance in security research that pairs stronger capabilities with better model alignment. The model powered the Linux kernel research we announced yesterday, and security teams at companies like Mozilla, Palo Alto Networks, Calif, and Cisco—alongside our own at OpenAI—are already putting it to work. GPT-6 Cyber is available to select Daybreak Red customers and we are expanding access over time. openai.com/index/introducing…
1
2
6
306
wait sol just reads your inbox and preps all the stuff and you just approve it???
🤝 Paid partnership
We shipped a BIG update to Sol (hellosol.app)! Sol is for people who have consequential work in their inbox. The most fun part tho - was watching the bets behind this update play out with early users: - 100+ new emails a day - is the norm for a LOT of people: founders, execs, sales, recruiters, consultants, ops. - To make things interesting - every email asks for something different: a decision, an approval, a meeting, a proposal. - The heavy lifting is in: arriving at the decision, doing the review, picking if/when to meet, creating the proposal. - Drafting the reply - is merely the outcome of that heavy lifting. That alone doesn’t solve the real problem. That’s where Sol is different. It works on every email as soon as it arrives, does the heavy lifting each one needs, and prepares it for your approval. You approve or edit. It learns each time and keeps getting better. When your inbox is filled with highly consequential work, you need an assistant that’s doing the heavy lifting, continuously. That’s what we’re building Sol to be - proactive about what you need, 100 times a day, every single day. Extending early access to this big update ✌️
1
2
6
193
was heading back from London and somehow ended up designing my own Flipper-inspired security board @ChihYang04 you’ve built something wild!
2
5
287
we’re just about ready to get the show started @ Dev Day Exchange London 🤩
2
4
215
they said no big launch wifi password was astra6.2 😭
OpenAI DevDay is going global. Starting this October, DevDay Exchange is bringing builders together to swap build notes, share real projects, and meet the teams building OpenAI tools in: Bengaluru Tokyo Seoul Berlin Paris London São Paulo Mexico City We're ready to see what developers around the 🌎 are pushing to prod.
4
2
84
11,970
who else is at devday london today fr? would love to link up if ur around 👀
we’re just about ready to get the show started @ Dev Day Exchange London 🤩
1
10
382
it's not that deep bro
5
76
Atlasly for Rhino v1. Agent that edits your actual model: height options, floor plates, area calculations and 185+ tools, powered by GPT-6 Luna. 128 benchmark tasks ≈ $0.96 in model usage. More coming soon.
2
2
6
371