Jay Patel (@jayypatel18) header
Jay Patel (@jayypatel18) profile picture

ex @jpmorgan turned founder • i let claude loose on real problems • builder at heart, I miss coding by hand

Joined March 2021
Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)Photo by Jay Patel (@jayypatel18)
grok bot called anthropic about my claude limits at 3am
361
616
120
24,068
1,224,888
update: it got me 2x limits. i don't want to know what it promised them
40
22
4,530
124,320
part 3 is out and tibo’s very helpful nitter.gq/jayypatel18/status/211…
grok bot started a bidding war between anthropic and openai for me at 3am
74
8,831
update 2: it's on the phone with openai now. someone named tibo picked up
2
145
15,200
part 3 is out and sam’s on the call nitter.gq/jayypatel18/status/211…
grok bot started a bidding war between anthropic and openai for me at 3am
1
21
7,464
Jay Patel retweeted
grok bot started a bidding war between anthropic and openai for me at 3am
12
7
3
111
27,701
grok bot started a bidding war between anthropic and openai for me at 3am
40
8,277
grok bot started a bidding war between anthropic and openai for me at 3am
12
7
3
111
27,701
update: my laptop is down to 8gb of ram. grok bot gave anthropic 24 hours to beat that
28
2,991
Jay Patel retweeted
Talth is live on Product Hunt today 🚀 It makes premium launch films for products. Write a brief, get a finished film, and edit anything later without starting over. Would mean a lot if you checked it out and left a comment with your honest thoughts 🙏
11
3
30
93,297
everyone's waiting for agi to cure cancer meanwhile grok bot is calling comcast for people. someone saved $500 on one call honestly this is the agi i wanted
7
15
815
mine has been making calls about my claude habit all week. part 3 tonight
5
348
claude can now read every transaction in your bank account it can't move a dollar. it can't stop you it just watches you order doordash at 3am and says "you're absolutely right"
Introducing Personal Finance in Claude, in beta in the US. Connect the accounts you choose, and Claude answers your money questions in everyday language, from your own balances and transactions. Access is read-only, and Claude can't move, manage, or invest your money.
4
9
781
openai now has two new models one refuses to hack anything the other found 948 ways into linux guess which one you're getting
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…
3
13
881
it's the first one. the other goes to a short list of security teams. openai rated it critical, the top of its own risk scale
4
328
gpt-6.2 isn't even out and my chatgpt dot already found out about claude
21
10
4
760
38,705
grok bot called anthropic about my claude limits at 3am
3
367
chatgpt images 2.6 makes gifs now so i had opus draw my chatgpt dot in code instead. mochi did not take it well
gpt-6.2 isn't even out and my chatgpt dot already found out about claude
3
7
1,315
gpt-6.2 astra, gpt-6.2 sol and a cyber model, maybe by friday the model picker is going to need its own model picker
OpenAIs GPT-6.2 could come as soon as Friday, per The Information. It will come in Astra, Sol and Cyber focused model.
2
8
935
openai's agents found and patched 948 linux kernel exploits in one run $1.2M in API costs. about $1,266 a bug bug bounty hunters just got undercut by an API
Our security researchers built an autonomous cyber vulnerability discovery, exploitation, and patching pipeline. In a single run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel at a total API cost of $1.2 million.
5
1
12
892
cat wu from claude code just joined openai my 83.1B claude code tokens are now in a custody battle
Some news: I’m joining OpenAI to lead product management for Core Products! I have a lot of respect for OpenAI for showing the world that scaling works, for bringing AI to over a billion people, and for believing great products are core to the mission. I’m excited to work with @thsottiaux, an incredible PM team, and everyone across OpenAI. Together, we’ll build agents that have your back and your team’s, and keep you in control, so you get more time for the people and work you love. Lots to learn, and lots to build. Please keep your feedback coming!
6
1
1
10
914