Vibinex Code-Review
A distributed process for reviewing pull requests.
Start new thread

Vibinex Code-Review - A distributed process for reviewing pull requests

by•
If your team misses delivery deadlines due to rework or resolved issues reappear in your product, then this is the tool for you. Vibinex automatically distributes the review process among developers at a code-hunk level for better quality and faster delivery.

Add a comment

Replies

Best
Maker
📌
Hey everyone! Super excited to finally launch our pull-request tool today. A few months back, as a software engineer myself, I set out to fix the most frustrating problems and elevate the development experience of all software engineers. Across several companies, the most frustrating yet critical part of leading a project was setting up the code review process and getting it adopted. With Vibinex, we are making that extremely frictionless and efficient. The core feature of Vibinex is the automatic distribution and assignment of code reviewing using git history. This changes the review process from being binary (approved or not approved) to continuous (e.g. 20% approved). Some ways in which this empowers developers: 1. Identify which PRs require your review: Going through 10 PRs is a hassle. But reviewing 2 PRs that require your attention is easily manageable. 2. Partially review a PR: You don't need to review every diff in a PR, just the code-hunks that are relevant to you. With Vibinex, you don't have to worry about miscommunication. 3. Better awareness: With this new process, you review everything relevant to you in lesser time. So you never have to worry about communicating outdated information about the product to your stakeholders. PRIVACY FIRST For doing all this, we don't even ask for access to your code. Our open-source Rust program (Code: ) runs in your CI pipeline, parses your git history to extract only the relevant metadata (making it impossible to reproduce your code from it), and then SHA-encrypts every word in it, including your email aliases, making them unreadable, and only sends that to our servers. When you log in using our browser extension (also, open-source: ), we get your email aliases and match their SHA-encrypted version with the data from our backend to show you the insights. COMPATIBILITY: Code hosting: Github and Bitbucket Browsers: Chromium-based (Google Chrome, Microsoft Edge, Brave etc.) Note: If your team uses a different code-hosting service or web browser and is keen on using our tool, we'd be happy to build the integrations to get you started.
Congrats on the launch! 😍 How do you ensure that the code review assignments are distributed fairly and evenly among team members?
Maker
We current assign code blocks based on the history of the code itself, i.e. the last editor of that piece of code. But in the future, we'll extract deeper insights (like active developers, module owners & production logs etc) to enhance this distribution algorithm further. If you have any more ideas, we'd love to hear them.
this already sounds like a great plan! Have you considered incorporating a weighted distribution algorithm that takes into account factors such as team members' expertise, workload, and availability? This would ensure a fair and balanced distribution of code review assignments based on the specific skills and capacity of each team member. I am also wondering if it would make sense to include some form of a rotation that would ensure every team member also gets to develop their skills in areas they are not proficient in (yet), so very much like ongoing training.
Maker
Yes, but it is a little further down in our roadmap since it is quite difficult technologically to extract workload and availability without integrating with a bunch of other data sources like the team's internal communication channel (Slack, MS Teams etc.) and their task management software (JIRA, Asana, Trello, ClickUp etc.). Skill and capacity are some things that we hadn't yet thought of. Thanks for the great suggestion. Regarding the ongoing training, we are actually contemplating several ideas for using pull requests and the code itself to minimize the number of "knowledge transfer" meetings that tech team leaders have to conduct. That is the 2nd biggest problem that we have identified among dev teams. I'd love to discuss this in more detail with you if you are open to it. Feel free to drop me an email at or choose any of the contact options from the footer of our website - they all reach me.
Ah, yeah, the struggle of integrations - one of our challenges at Collato as well :) But it sounds like you've got a solid plan in place. And finding ways to reduce those knowledge transfer meetings is a brilliant idea! Your tech team will thank you for it. Going back to the rotation, I was just thinking what if team members were chosen to take turns being responsible for code reviews to promote knowledge sharing and foster a collaborative environment? Getting over my head here hahaha but will stop now and drop you an e-mail. Let's keep the conversation going!
Looks like a real hassle saver for developers. Kudos team!
and are two of the most talented and dedicated people I know. I've been following their progress for a while now, and I'm so excited to see their hard work come to fruition. Congratulations team!
Maker
Thank you so much, Sarvesh, for your support!!
This product will surely solve some real issues
Congratulation on you launch.. and best of luck.
Awesome work guys! Keep it up! Godspeed!
congratulations on the launch!
Congratulations on the launch and ! A continuous review process instead of discreet seems like such an obvious thing to have.
All the best
123
Next