DVC Extension for VS Code — Track machine learning experiments right in your IDE

Hey PH! I’m co-founder at Iterative and we are a team that builds and maintains DVC – a popular open source tool for ML. Today we are excited to share a DVC VS Code extension that turns the most popular IDE into a fully functioning ML experimentation platform: ✅ Run, track, and compare your experiments, metrics and params ✅ Queue multiple experiments and run them in parallel to try different ideas faster ✅ View deep learning metrics in real-time ✅ Make experiments reproducible with data, code, metrics, hyperparameters and models tracked together And no additional moving parts! No need to run any servers, no dependency on third-party SaaS solutions, etc. *Our backstory.* We started the company 6 years ago with a big vision to bring best practices, collaboration, and rigor into ML / DS workflow, while also providing the best experience to ML and DS folks. Large tech companies had heavy internal DS collaboration platforms for many years, clearly those tools were not useful and impossible to use outside. My co-founder Dmitry created DVC as an open source project, that is lightweight, requires only Git and CLI and brings reproducibility to ML projects. While DVC is a cool idea and became a very popular ML tool that is used by thousands of ML teams, we quickly realized that the CLI, Git-like interface can be a challenge as well. While giving really great guarantees similar to Git - immutability, codification, etc, etc, it was clear that we also need to meet our end users where they are - notebooks, Python scripts, VS Code, etc. Last year we made a lot of progress and this VS Code extension wraps all of it up into a really nice UI and UX within the most popular platform at the moment. Try it out and let me know what you think! 🤗

Kudos to our amazing team for the successful launch of the DVC VSCode extension! 🌟🚀 Your creativity and effort have shaped this revolutionary tool!

🎉 Kudos on the launch! The DVC Extension for VS Code is a game-changer for AI developers wielding DVC tools. I'm thrilled about the lightning-fast evolution and the shiny new features gracing the DVC Extension. It's not just about parameters and metrics comparison anymore – we're talking live metrics plots and images! 📊✨ The user interface has leveled up, flexing its muscles with newfound versatility and potency. Hold onto your hats, folks! GitOps ML just got a whole lot friendlier with this gem! 🚀🤖🪙

Really proud of how far this extension has come! It's been a pleasure working on this with my fellow team members!

I am proud to see how far the extension has come after all the hard work the team has put into it. I'm pleased that people use it and see it as an essential tool for their workflow.

Congratulations to the team for all the hard and effective work they have put into the extension to improve the experience for all our Community members. The DVC extension for VS Code made a giant leap forward in the last year. Kudos to our whole team for their work and support!

I am super excited about the launch of DVC VS Code extension! 🎉 The team has worked very hard to make the popular IDE a powerful platform for ML experiments. It is one step forward towards making ML accessible to everyone!

The DVC extension for VS Code makes it so much easier to work on your DVC projects _near_ your code! Shoutout to Matt, Stephanie and Julie who have been working hard on making the experience even greater - good luck!

Proud to launch the DVC extension for VS Code on Product Hunt today! The extension has reached a level of maturity where it is now my go-to environment for model development. Hope you find it useful!

Congrats on the launch, team!

Aha! I worked with folks at DVC back in 2020, on the SaaS version of this, DVC Studio. They're doing something absolutely great with the extension, providing intuitive GUI tools for folks who are working on machine learning problems, making it easy to onboard new folks into teams, and easy to collaborate with other teams (eng, product, etc etc). Congratulations on launching the extension, and on the whole series of launches with DVC 3.0! 6 years is a very long time in startup years. Stay strong and focused. @shcheklein @fullstackml 👏🏼 ❤️‍🩹

Hey there, PH! 👋 I had the opportunity to take part in the journey of DVC. What a ride it's been! As a contributor, it has been great to see the extension growing and evolving! Been around for a while… from experiencing what was a hard config from early versions to the “aha” moment after the 0.5.3 release introducing UI for easily configuring DVC…Happy to see how it has been moving from CLI to a more extension-friendly design and usage over time… 𝚍𝚟𝚌 𝚎𝚡𝚙 𝚜𝚑𝚘𝚠 anyone? Besides, I have had the opportunity to contribute to some issues and discussions, as well as reporting bugs that were related to machine learning common experimentation mental models, while learning about how to contribute in the Open. The funniest bugs were those corner cases in which some of the components disappeared 🙂 Looking forward to seeing how this tool mutates and evolves into something fully accessible to Data Scientists!

Congrats DVC VSCode team, now everybody can leverage experiment tracking and comparisons from VSCode, so fantastic! 🫶🔥

Congratulations on your launch to Product Hunt. It'll definitely save a tremendous amount of time we spent on experiments for sure.

Now I have a legitimate case of moving from PyCharm to VSCode. What an amazing idea and execution! Congrats on your launch, guys!

Congrats on the launch! Does this integrate directly with the experiment tracking?

Congratulations on the launch of the DVC extension for VS Code! This is such a handy tool for ML developers looking to streamline their workflow. Excited to see how it transforms the way we manage and track experiments. Kudos to the team! 🚀🎉

As a big fan of open-source projects, I've known about DVC for many years, and many data scientists I've spoken to use it in their work. It's great to see that the product keeps growing and benefiting the community.

Easy to use, powerful and open source. DVC VSCODE extension is one of the best tools within the ML ecosystem without any doubt. Please try it if you want to upgrade your ML engineer capabilities. Congratulations for the great work!

Congratulations on the launch! A suggestion I have is to consider integrating a feature that allows for collaborative work on experiments. Given the rise of remote work and distributed teams, this could be a huge boon for ML and DS teams working together on projects.