suchintan

Have you ever given Skyvern ( https://github.com/Skyvern-AI/skyvern ) a try? I'd love to hear your opinion…

That would be amazing!!

Very cool. The benchmark can be found here if you want to take a look at it: https://github.com/Halluminate/WebBench …

We definitely plan to expand it. I want to get to ~10,000 for a reasonable benchmark. 15 blew my mind -- it's too easy to overfit that dataset

Definitely. It gives you a lot more control on which models you want powering these browser agents, which is the important part

Hey HN, we were playing around with MCPs over the weekend and thought it would be cool to build an MCP that lets Claude / Cursor / Windsurf control your browser: https://github.co…

I've seen people use semantic versioning to version APIs and SDKs https://semver.org/ We at Skyvern are still doing patch versions only…

I wonder if more companies should open source their eval model outputs alongside the eval results We tried doing that here at Skyvern (eval.skyvern.com)

Definitely need a newer benchmark. I couldn't find where browser-use published their run results (expected to see it here https://github.com/browser-use/eval ) We went ahead …

I'd love to chat to see how we can help! Here's my email: suchintan@skyvern.com We're working on 2 major improvements that will get cost down at scale: 1. We're building a code gen…

This is a great point -- the example we chose was meant to be a consumer example that we could relate with.. however a similar example exists for the enterprise which may be more interesting Let'…

It creates a feedback loop for the LLM to help it correct hallucinations or misunderstandings about how it's planned actions actually played out!

This is very interesting, thank you for sharing it What are your opinions on lists like this? https://github.com/mmccaff/PlacesToPostYourStartup …

Yep. Makes a lot of sense. Your side project is super noble -- thank you for that

This is really cool. Do you have any interest in helping people auto apply to them? We can help you set it up with a really simple API call via Skyvern ( https://github.com/Skyvern-AI&#…

we started this last year. Back then, autogpt was more of a prototypical framework -- I'm sure it's improved dramatically since then. We ran into too many issues while developing with it (ho…

Yep. This is totally fair feedback -- we're still a super early product and haven't had a chance to optimize the phone experience.. largely because it's tough to see the magic from the …

Yes -- in theory. You'd need to use our workflows feature to get that set up and chain a few tasks together to collect that information!

Great to see you on here Anton! Just curious, how do you differentiate from other open source competitors like n8n and Active pieces?

You're making some really good points here 1/ the current prompt + payload structure is definitely on the complicated end of the spectrum, but we've found that we can use an LLM to help…

1. Yes absolutely. But the issue is a little bit more nuanced than that. Websites without APIs don't have them for one of two reasons: (1) They want to protect their data (LinkedIn) or (2) can&#x…