Janet AI Thinks Jira Is Broken Beyond Repair, So They Built a Replacement From Scratch

The Macro: Jira Is the Software Everyone Uses and Nobody Loves

I have never met an engineer who enjoys using Jira. I have met thousands who use it daily. That gap between adoption and satisfaction is one of the largest in enterprise software. Atlassian turned Jira into a $60 billion company not because the product is beloved but because the switching costs are terrifying and the alternatives have historically been worse in different ways.

Linear changed the conversation. It proved that ticket management could be fast, opinionated, and pleasant to use. Linear is what Jira would be if you deleted 80% of the features and made the remaining 20% beautiful. Shortcut (formerly Clubhouse) took a similar approach. Height added AI features early. Plane went open-source. The project management space has more competitors than ever, and most of them are iterating on the same fundamental model: humans create tickets, humans update tickets, humans close tickets.

The problem with that model is that it requires discipline, and engineers are not disciplined about ticket management because ticket management is not their job. Their job is writing code. Tickets are overhead. The result is predictable: boards full of stale tickets, half the actual work happening in Slack threads that never become tickets, decisions made in meetings that nobody logs anywhere, and a project manager spending hours every week manually updating statuses that are already out of date.

What if tickets just happened automatically? What if the conversation in Slack about a bug became a ticket without anyone creating it? What if the meeting where someone said “we need to refactor the auth service” became a ticket with an assignee and a priority? What if tickets moved to “done” when the PR merged, without anyone dragging a card? That is the AI-native version of ticket management, and it is a genuinely different product from everything else on the market.

The Micro: Founding Engineers Who Lived the Pain

Adhit Sankaran is the co-founder and CEO. He was a Founding Engineer and GenAI Engineering Lead at a YC S23 company, interned twice at a large tech company, and has a Master’s in Computer Science from Cornell with a focus on generative AI research. William Barthell is the co-founder and CTO. He was Employee number one at a crypto startup where he designed algorithmic trading mechanisms and helped secure over $3 million in seed funding. He has a Bachelor’s in CS from UW-Madison. Both are technical founders who have been the first engineers at startups before, which means they understand the operational pain of ticket management from the inside.

Janet AI (Y Combinator Summer 2025) is an AI-native ticket management system that integrates with Slack, Discord, SMS, email, and meetings to automatically create and organize work items. The core product does several things that existing tools do not. It listens to your communication channels and creates tickets when it detects actionable work. It auto-tags, detects duplicates, and suggests assignees. When a PR merges on GitHub that resolves an issue, the ticket updates itself. You can communicate with the person who reported a bug directly through the platform without switching to Slack or email.

The pricing model is interesting: usage-based with no monthly subscription. The free tier includes $100 in credits, unlimited tickets, unlimited users, and unlimited projects. This is a bet that the product is sticky enough that teams will naturally consume credits as they scale. It also removes the per-seat pricing friction that makes tools like Jira and Linear expensive for large teams.

The product also works with coding agents like Cursor and Claude Code, which is forward-looking in a way that matters. As more code gets written by AI, the connection between “what was decided” and “what was built” needs to be automated. If your coding agent resolves an issue, Janet should know about it and update the ticket. That feedback loop is where ticket management intersects with the broader AI-assisted development workflow.

The AI chat assist feature is available on the dashboard, in Slack, and via SMS. This means a product manager can ask “what is the status of the auth refactor?” and get an answer without opening the project board. That is a small thing that saves a surprising amount of time.

The Verdict

Janet AI is making a big bet: that the entire paradigm of manual ticket creation and management is going away. I think they are right, but the timing question is tricky. Linear proved that a better Jira can win market share. Janet is arguing that a different kind of product, one where tickets are a byproduct of work rather than a prerequisite for it, is the next step.

The biggest risk is the cold start problem. Janet works best when it is connected to all your communication channels, your GitHub repos, and your meeting tools. Getting a team to connect all of those integrations requires trust, and trust with a new startup is hard to earn. Linear won early adopters by being fast and beautiful. Janet needs to win early adopters by being invisible and accurate. If the auto-created tickets are wrong or noisy, teams will turn off the integration in a week.

Thirty days, I want to see the false positive rate on auto-created tickets. If every Slack conversation about lunch generates a ticket, the product is dead. If the signal-to-noise ratio is above 80%, teams will adopt fast. Sixty days, the question is whether product managers or engineers drive adoption. Engineers want fewer tickets. Product managers want more visibility. Janet needs to satisfy both. Ninety days, I want to know if teams are actually deleting their Linear or Jira accounts, or if Janet is running alongside them as a secondary system. Replacing Jira is a fifteen-year problem. But every year, the case for the replacement gets stronger, and Janet is the most compelling version of that replacement I have seen.