I switched over to Superset from Conductor a few months ago and haven't looked back - it's really nice to be able to use the native Codex/Claude Code TUIs without any of the bloat
yes! we've found that reading changes this way make it very easy to separate the important stuff from the unimportant stuff, and we're thinking of ways to make that more visible in the UI
And yeah, I think number 5 on your list is particularly interesting - juniors will develop much slower if they don't go through the struggle of understanding implementation
We're hoping that our tool can help make that easier
Well I think the bigger issue here is that GitHub is too decoupled from an issue tracking system - its a lot of manual overhead to constantly have to keep in sync. Linear does an ok job but far from ideal
Yeah we've been thinking a lot about this! Switching back and forth between local and GitHub is also a frustration of ours so we want to think of ways to unify the experience
Good question, we're more focused on the first aspect you mentioned of "does the code you wrote make sense on a high level" rather than "did you miss something" - I think tools like Greptile and Codex are coming out with new security review features that are more suited towards that use case
Great point and that's our takeaway too from talking to many users. We're exploring ways for Stage to possibly tailor the review flow to each specific PR/user preferences. Would love to hear any ideas if you have any!
Yep that seems to be a common sentiment! Definitely some interesting areas around capturing agent conversations in commit history that could make this possible
Good point, we've heard from users that there's a lot of fatigue of switching between local and GitHub all the time so we're thinking of ways to unify that experience
Thanks for sharing this, I do agree with a lot of what you said especially around trust around what its actually telling you
For me, I only run into problems of an agent misleading/lying to me when working on a large feature, where the agent has strong incentive to lie and pretend like the work is done. However, there doesn't seem to be this same incentive for a completely separate agent that is just generating a narrative of a pull request. Would love to hear what you think
I agree with that - with Stage we're not trying to replace reading code with AI summaries, but rather guiding the reviewer through reading code in the way that makes most sense and coming away with the best understanding
Totally get that, still something we're actively talking about!
Sort of related to that, we've been thinking a lot about the future of code review for OSS. Its clear with Cal.com going closed source that something needs to change. Would love to hear any thoughts you have
Can't wait to see what else you guys cook up!