"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints."
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password As a longtime 1Password user and software engineer, this article seriously undermines my confidence.
For security-critical software, AI should assist vulnerability discovery, with every finding rigorously validated by senior engineers, not drive development velocity.
I’d rather hear about how you are achieving reliability standards comparable to avionics or medical device software than excitement about AI-assisted coding.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password So where do I switch to that does not use LLMs for this? So sad that so much once great software gets worse these days.
-
I dropped them, after a decade, and have gone with Apple's Passwords. So far, so good. I also added Uplock for financial info.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password Shame my subscription just renewed a couple months back, but I guess that means I have ample time to find a better platform.
This is always bad, but extremely bad in a tool guarding my valuable secrets. I'm out.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password I guess I need to find a replacement password manager asap. Well done.
-
I got my wife to make the move. It was relatively painless. She does not like info tech at all.
-
I got my wife to make the move. It was relatively painless. She does not like info tech at all.
Also easy because all of our devices are Apple
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password Not cool. Not cool at all.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password I’m not a fan of this AT ALL. I’LL be looking at alternatives for my team that won’t be relying on AI slop. Kind of explains why your apps have been kind of buggy recently.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password who needs a $5 wrench?
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password Please write a blog post on the feedback you got on this post. Was it universally negative? Did you learn anything?
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password Ah, so this deconstruction is why I have been getting to be almost daily system issue emails across the cloud stack which say they are for "scale issues" it's really you teasing out pieces then having to tape over the leaks?
There was an implicit trust when you dragged us from Dropbox/iCloud local sync to your cloud. You've expanded your product set outside your historical core business and now money and scale are meeting the monolith. I hope us customers survive.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password hint: the ideal "agent to human ratio" is 0. seeing a company jeopardise its reputation like this just to chase a trend is very disappointing.
-
@1password As a longtime 1Password user and software engineer, this article seriously undermines my confidence.
For security-critical software, AI should assist vulnerability discovery, with every finding rigorously validated by senior engineers, not drive development velocity.
I’d rather hear about how you are achieving reliability standards comparable to avionics or medical device software than excitement about AI-assisted coding.
@chadpod @1password So well said!
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password PLEASE could you GA Passkeys for logging into 1Password....
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password Alright, it was nice knowing you. Take care.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password Oh FUCK no.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password WTF? Just recently I decided to stay and accept your price increase, but THIS is definitely not acceptable.
Just canceled our family plan.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
@1password In retrospect I'm so glad I didn't shell out for a 1Password account. The product looked solid, but clearly the company cannot be trusted with sensitive information.
-
"The pattern that works is using agents to produce deterministic artifacts, then forcing execution through those constraints." Tido Carriero, VP of Engineering at Cursor.
At 1Password, we applied agentic tooling to B5, our multi-million-line Go monolith, to help plan and execute a production refactor. Here's what we learned: https://1password.com/blog/what-we-learned-using-ai-agents-to-refactor-a-monolith
Actually, a good summary of the lessons. From the business and engineering perspectives, I have a few questions; How do you measure ROI? When is it advantageous for engineers to leverage LLMs, when would it be more beneficial to hire a new FTE?
Finally, how do you maintain engineer motivation, especially when LLMs can handle a significant portion of their work? And how do you ensure a consistent influx of junior engineers while also fostering their continued learning?
At the end of the day, LLMs are trained by data created by engineers. No engineers left == no data for LLMs to train.
Compared to others in the comments, I'm actually happy to see how you think about using LLMs within the organization.
