thedg

Hey HN, when I was a PM at Cloudflare 7 years ago, whenever we’d test our site on mobile, it was super annoying to make Jira tickets: I’d take a bunch of screenshots on my phone, email them to myself, then go one by one and make Jira tickets for them.

Now 7 years later and this is actually still the way people do mobile QA. It takes ages to create 1 ticket let alone 10. How is there no tool for this yet!

We just pushed an app to the App Store that lets you record your screen and make a Jira/Linear/etc ticket right from your phone, and the ticket auto-includes device/OS level details like battery, memory, etc to help engineers repro.

https://apps.apple.com/us/app/jam-fix-bugs-faster/id64690372...

I would love to hear what you think, how we can develop it more to make reporting bugs from your phone practically instant. We’re looking at ways to capture console logs next. Curious what else you think would be good to add.

PS we are hiring - especially Android engineers to build this for Android too - so if you are interested, would love to hear from you: jam.dev/careers

Today we added Gemini 2.0 to our web debugger (jam.dev) to let it analyze the browser session and DevTools and write up a bug report when there's an error.

It even writes repro steps (and adds timestamps to link each repro step to the exact moment in the session recording playback). Plus, it's a lot more concise than most people reporting bugs, so engineers can understand immediately what is going on without having to read a lot of text or decipher a confusing ticket :)

You can see the demo here: https://x.com/jamdotdev/status/1887523689518014891

One cool feature: once you `tab` to accept Gemini's AI-written bug report, it will log it right in your issue tracker (so you never have to open Jira to log a Jira again). (We integrated with Jira, Linear, and 7 others)

Why Gemini 2.0? We initially chose Gemini because of its native ability to process videos (which other LLMs don’t have, so it would require a lot less pre-processing for us and avoid adding a small amount of latency). And then we enabled Gemini 2.0 while it was in experimental mode. It turns out Gemini 2.0 is pretty good at logging bugs. It's fully writing 27% of the early access users bug reports, which we are pretty happy to see.

You can try it out for yourself: jam.dev/ai I’d love to hear your thoughts, my email is dani@jam.dev – reach out anytime. We want to make this a lot more fully featured – and would love your input where you'd like to see this go.

Definitely we think humans have a major role to play in testing software, but compiling the debug logs and writing up the summary and repro steps for engineers seems like something AI excels at and people are a lot more imprecise at (and that imprecision causes a lot of wasted effort today on engineering teams).

By the way, we’re hiring engineers and if you're really into AI-powered devtools and debugging, would love to chat: jam.dev/careers

Link: jam.dev/ai Demo: https://x.com/jamdotdev/status/1887523689518014891

People are terrible at writing bug reports.

It’s insane that bug reporting is still done by hand. That’s how you end up with hours of back-and-forth for a 5 min fix. "What browser?" “Is it still happening?”

But this is a problem you can fix with tools. Humans are inherently error-prone when it comes to reporting bugs, but computers aren’t. Imagine a tool that plugs into devtools and grabs: session recording + network logs + error traces + AI that writes a summary and repro steps.

A few years ago, we built a tool that captures all the information you’d get in a HAR file plus screen recording and packages it into a link for engineers. (Like a screen recorder plus DevTools). It’s called jam.dev and people use it to capture more than half a million bugs each month now (which we are really excited about!)

Today we added an AI inside that analyzes all of the session data and errors/traces in the console and network logs and writes a concise description of the bug and lists out the exact repro steps, complete with timestamps mapped to the screen recording. It’s built with Gemini 2.0 which we used while in experimental mode. We chose Gemini because of Gemini’s native ability to process videos (which other LLMs don’t have).

I’d love for you to try it out and let me know your thoughts. My email is dani@jam.dev and I hope to hear from you!

I think people are going to have a large role in testing software for a long time (as long as humans are the users, humans need to be testers too), but what doesn’t need to be manually done is collecting debug information and writing it up into a ticket. Like Kasparov and Deep Blue, the strongest collab is human + computer paired together –– humans to their strengths, computers to theirs.

By the way, we’re hiring engineers and working on building AI for debugging, de-duplication and auto-resolution and if these are problems that excite you, we’d love to chat: jam.dev/careers

It’s insane that bug reporting is still done by hand. People are terrible at writing bug reports. That’s how you end up with hours of back-and-forth for a 5 min fix. "What browser?" “Is it still happening?”

But this is a problem you can fix with tools. Humans are inherently error-prone when it comes to reporting bugs, but computers aren’t. Imagine a tool that plugs into devtools and grabs: session recording + network logs + error traces + AI that writes a summary and repro steps.

A few years ago, we built a tool that captures all the information you’d get in a HAR file plus screen recording and packages it into a link for engineers. (Like a screen recorder plus DevTools). It’s called jam.dev and people use it to capture more than half a million bugs each month now (which we are really excited about!)

Today we added an AI inside that analyzes all of the session data and errors/traces in the console and network logs and writes a concise description of the bug and lists out the exact repro steps, complete with timestamps mapped to the screen recording. It’s built with Gemini 2.0 which we used while in experimental mode. We chose Gemini because of Gemini’s native ability to process videos (which other LLMs don’t have).

I’d love for you to try it out and let me know your thoughts, my email is dani@jam.dev.

Try it: jam.dev/ai Demo: https://drive.google.com/file/d/1g1RwoC8ITTTWR6BoIrgi_1YD6iD...

I think people are going to have a large role in testing software for a long time (as long as humans are the users, humans need to be testers too), but what doesn’t need to be manually done is collecting debug information and writing it up into a ticket. Like Kasparov and Deep Blue, the strongest collab is human + computer paired together –– humans to their strengths, computers to theirs.

By the way, we’re hiring engineers and working on building AI for debugging, de-duplication and auto-resolution and if these are problems that excite you, we’d love to chat: jam.dev/careers

Hey HN! 4 years ago, an ex-Cloudflare team and I started building Jam.dev –– trying to make bug reporting a lot less painful.

It's crazy that in 2024, the state of the art way to debug a reported issue is to hop on a call. How is there no tooling?!

So we built a tool we needed back at Cloudflare: fast forward to today and it's used by 150k people all over the world.

But it still only solved when PMs or QA reported bugs to engineers. We still hadn't quite figured out how to help users –– usually the worst bug reporters of them all –– report bugs better.

There are tools out there for session recording, but then to find the exact moment you need, you need to scrub through a 45 min video of someone's entire session. It's 1) creepy 2) not efficient.

So we kept thinking.

Earlier this year, an engineer on our team prototyped something we could use in our own customer support chats. It's a plugin that lets support agents request a screen recording (with dev logs) from the customer.

As we started to use it ourselves, we realized it was saving us from having to hop on calls with customers and ask them a lot of questions to figure out what was happening.

While recording, we automatically capture:

- Console logs

- Network requests (fully inspectable, w/ copy as cURL for local repro)

- Timing waterfall

- Session metadata (User ID, etc) & system status (battery, etc)

- Device/browser/OS info

There’s some extra debug tooling in there for websockets and graphql too (we love/hate graphql, but that’s besides the point.)

You can see an example of what the captured data looks like here: https://jam.dev/c/cff872c4-0a40-4bc2-a35a-9623cc4926a6

It’s a lot more private than the session recording tools out there because it’s only recording when the user explicitly starts/stops the recording, and not the full session.

I hope you’ll check it out and let us know what you think: jam.dev/customer-support

The worst part of debugging customer issues is not even the debugging, it’s just figuring out what the user is even talking about. It’s so crazy that there’s not much tooling out there to make that a lot faster, even in 2024. We really want to build something that makes this part of the job suck less, and we'd love to know what you think and how we can make it even better.

My email is dani@jam.dev, feel free to reach out anytime.

By the way, we’re hiring engineers and if making debugging a lot more efficient is a problem that excites you, we’d love to chat: jam.dev/careers

Hey HN, wanted to share a new collection of dev utilities that are ad-free, client-side (nothing leaves your device), with a command+k menu for fast search.

github.com/jamdotdev/jam-dev-utilities

Why? The sites we use every day for data conversion (e.g. PNG to data URI) are cluttered with ads, one of those mildly annoying things engineers deal with while writing code.

This came up at a recent team dinner, and we got interested in creating nicer, modern, uncluttered versions of these tools. Since then, every time we needed a util, we added it to our stack. Lots more tools still to add, but so far we’ve got 15 like JSON converters to/from CSV, YAML, etc.

You can check them out at jam.dev/utilities. It’s built with next.js, tailwind css, and shadcn/ui. It’s ad-free, open source, and all runs client side – data never leaves your computer. There’s a command K search powered by cmdk.paco.me.

Are there any other tools that you wish existed or are there any other tools that like you've made for yourself like this?

By the way, we’re hiring engineers and if you like building dev tooling, that’s our jam and we’d love to chat: jam.dev/careers