6 ms·

GitLab was rather quick to board the AI hype train this year. Fine - they don't want to look like they're behind the curve, have to please investors, and whatev

GitLab was rather quick to board the AI hype train this year. Fine - they don't want to look like they're behind the curve, have to please investors, and whatever else. But IMO there are plenty of other places to add real value across the GitLab product with AI/ML features. Here, it just looks like they saw GitHub do something and felt a need to copy it. But two years late, and worse. As a longtime GitLab user (and onetime contributor!), I'm a bit disappointed they're spending so much time on this. I think they're just too far behind.

dnsmichi 3y ago
GitLab team member here. Thanks for your feedback. > But IMO there are plenty of other places to add real value across the GitLab product with AI/ML features. True, and after starting with ML experiments, the product and engineering teams have been working on new features for entire DevOps lifecycle. All AI workflows on the DevSecOps platforms are described in the GitLab Duo announcement blog post https://about.gitlab.com/blog/2023/06/22/meet-gitlab-duo-the-suite-of-ai-capabilities/ https://about.gitlab.com/blog/2023/06/22/meet-gitlab-duo-the... and website https://about.gitlab.com/gitlab-duo/ https://about.gitlab.com/gitlab-duo/ I'll share a few highlights that I am personally excited about - Explain and help fix security vulnerabilities. From my personal experience, I often find CVEs hard to read, especially when I am not the author of the code to fix. Getting help from AI can reduce entry barriers and make development for efficient. Security is everyone's responsibility these days. This follows the AI assisted feature to explain code in general. "What does this magic loop with memcpy do?" might not stay magic anymore, easing the path to code refactoring, improving performance, and reduce the resource usage footprint. - Summarize issue comments. Feature proposals or bug analysis can have long comment threads that require reading time. AI will help get the gist and better contribute to what has been discussed. - Summarize MR changes, to avoid reading long change diffs. This helps with faster (code) review cycles. I tested it this week with an MR for our handbook in https://gitlab.com/gitlab-com/www-gitlab-com/-/merge_requests/126426#note_1449388720 https://gitlab.com/gitlab-com/www-gitlab-com/-/merge_request... I'd also like to see AI helping fix CI/CD pipelines fast. Proposal in https://gitlab.com/gitlab-org/gitlab/-/issues/386863 https://gitlab.com/gitlab-org/gitlab/-/issues/386863 I shared some thoughts in a new talk "Observability for Efficient DevSecOps Pipelines", slides in https://go.gitlab.com/VDAvMw https://go.gitlab.com/VDAvMw (GitLab blog post coming soon, https://gitlab.com/gitlab-com/www-gitlab-com/-/issues/34296 https://gitlab.com/gitlab-com/www-gitlab-com/-/issues/34296) Additionally, I learned some new ideas at Cloudland last week, regarding product owner requirements list verification, and end-to-end test automation with AI. Need to create feature proposals :-) > As a longtime GitLab user (and onetime contributor!), Thanks for contributing. I'd like to invite you to share your ideas about AI features across the platform :) When you look at the DevOps lifecycle (image in https://about.gitlab.com/gitlab-duo/ https://about.gitlab.com/gitlab-duo/) from plan/manage to create, verify, secure, package, release, deploy, monitor, govern - where do you see yourself, and where do you spend the most time in? Second question: Which process feels the most inefficient? After identifying answers to the questions, please check the AI features https://docs.gitlab.com/ee/user/ai_features.html https://docs.gitlab.com/ee/user/ai_features.html and/or open new feature proposals for GitLab https://gitlab.com/gitlab-org/gitlab/-/issues/new?issuable_template=Feature%20Proposal%20-%20lean https://gitlab.com/gitlab-org/gitlab/-/issues/new?issuable_t... You can tag @dnsmichi so I can engage with your ideas. Thanks!
jgalt212 3y ago
> have to please investor I think it's more an internal itch. For a persistent money loser, the stock in holding there pretty well. https://finance.yahoo.com/quote/GTLB?p=GTLB&.tsrc=fin-srch https://finance.yahoo.com/quote/GTLB?p=GTLB&.tsrc=fin-srch

dnsmichi 3y ago

GitLab team member here. Thanks for your feedback. > But IMO there are plenty of other places to add real value across the GitLab product with AI/ML features. True, and after starting with ML experiments, the product and engineering teams have been working on new features for entire DevOps lifecycle. All AI workflows on the DevSecOps platforms are described in the GitLab Duo announcement blog post https://about.gitlab.com/blog/2023/06/22/meet-gitlab-duo-the-suite-of-ai-capabilities/ https://about.gitlab.com/blog/2023/06/22/meet-gitlab-duo-the... and website https://about.gitlab.com/gitlab-duo/ https://about.gitlab.com/gitlab-duo/ I'll share a few highlights that I am personally excited about - Explain and help fix security vulnerabilities. From my personal experience, I often find CVEs hard to read, especially when I am not the author of the code to fix. Getting help from AI can reduce entry barriers and make development for efficient. Security is everyone's responsibility these days. This follows the AI assisted feature to explain code in general. "What does this magic loop with memcpy do?" might not stay magic anymore, easing the path to code refactoring, improving performance, and reduce the resource usage footprint. - Summarize issue comments. Feature proposals or bug analysis can have long comment threads that require reading time. AI will help get the gist and better contribute to what has been discussed. - Summarize MR changes, to avoid reading long change diffs. This helps with faster (code) review cycles. I tested it this week with an MR for our handbook in https://gitlab.com/gitlab-com/www-gitlab-com/-/merge_requests/126426#note_1449388720 https://gitlab.com/gitlab-com/www-gitlab-com/-/merge_request... I'd also like to see AI helping fix CI/CD pipelines fast. Proposal in https://gitlab.com/gitlab-org/gitlab/-/issues/386863 https://gitlab.com/gitlab-org/gitlab/-/issues/386863 I shared some thoughts in a new talk "Observability for Efficient DevSecOps Pipelines", slides in https://go.gitlab.com/VDAvMw https://go.gitlab.com/VDAvMw (GitLab blog post coming soon, https://gitlab.com/gitlab-com/www-gitlab-com/-/issues/34296 https://gitlab.com/gitlab-com/www-gitlab-com/-/issues/34296) Additionally, I learned some new ideas at Cloudland last week, regarding product owner requirements list verification, and end-to-end test automation with AI. Need to create feature proposals :-) > As a longtime GitLab user (and onetime contributor!), Thanks for contributing. I'd like to invite you to share your ideas about AI features across the platform :) When you look at the DevOps lifecycle (image in https://about.gitlab.com/gitlab-duo/ https://about.gitlab.com/gitlab-duo/) from plan/manage to create, verify, secure, package, release, deploy, monitor, govern - where do you see yourself, and where do you spend the most time in? Second question: Which process feels the most inefficient? After identifying answers to the questions, please check the AI features https://docs.gitlab.com/ee/user/ai_features.html https://docs.gitlab.com/ee/user/ai_features.html and/or open new feature proposals for GitLab https://gitlab.com/gitlab-org/gitlab/-/issues/new?issuable_template=Feature%20Proposal%20-%20lean https://gitlab.com/gitlab-org/gitlab/-/issues/new?issuable_t... You can tag @dnsmichi so I can engage with your ideas. Thanks!

jgalt212 3y ago

> have to please investor I think it's more an internal itch. For a persistent money loser, the stock in holding there pretty well. https://finance.yahoo.com/quote/GTLB?p=GTLB&.tsrc=fin-srch https://finance.yahoo.com/quote/GTLB?p=GTLB&.tsrc=fin-srch