Oki Reads Your Team's Work So You Don't Have To

The Macro: Status Meetings Are a Tax on Everyone’s Time

I have a theory that roughly 30% of all engineering manager hours are spent trying to figure out what their teams actually did last week. Not because the work is invisible, but because it’s scattered across six different tools. The commit history tells part of the story. Jira tells another part. Slack tells a third, usually contradictory, part. And then someone puts together a Google Doc that nobody reads.

The status update problem is real and it’s expensive. A team of ten engineers with a weekly standup and a weekly status meeting is burning 20 person-hours a month just on “what are you working on?” rituals. That’s before you count the async Slack threads where managers are chasing context because the standup didn’t actually cover enough detail.

There’s been a wave of tools trying to solve this. Geekbot automates standups in Slack. Range tried to build a “team success platform” (whatever that means). Lattice bolted status updates onto their HR product. Linear has solid project tracking but it only knows about tickets, not code or conversations. None of these tools actually read your team’s work. They ask people to self-report, which means they inherit all the biases and gaps of self-reporting.

The AI angle here is obvious and, for once, actually well-suited to the problem. LLMs are good at reading lots of text, finding patterns, and summarizing. That’s literally what a status update is. The question is whether anyone can build a product around that capability that engineering leaders actually trust enough to replace their existing rituals.

The Micro: Two Berkeley Grads Watching Your Commits

Oki connects to your team’s tools (GitHub, GitLab, Slack, Jira, Notion, Linear, Google Docs, Microsoft Teams) and generates weekly AI reports for every team. The reports are interactive, with clickable source citations so you can verify anything the AI tells you. It auto-categorizes teams and projects, and there’s a contributor stress monitoring feature that flags when someone appears blocked or overloaded.

The founding team is Luofei Chen (CEO) and Aayush Tyagi, both out of UC Berkeley. Luofei did the M.E.T. dual degree in engineering and business. Aayush spent five years as an Android developer at Robinhood, which means he’s seen what engineering management looks like at a company that scaled fast and chaotically. They’re part of YC’s Winter 2025 batch, working with Harj Taggar as their primary partner. Two-person team, which is lean even by YC standards.

The product positioning is smart. They’re not trying to replace your project management tool. They’re not asking engineers to change any behavior. The pitch is: keep working the way you work, and Oki will tell your manager what you did. That’s a much easier sell than “migrate your whole team to our platform.”

The compliance angle is interesting too. SOC II Type II and HIPAA compliance, zero training on customer data, and on-premise deployment options. That’s unusually mature security posture for a two-person startup and suggests they’re going after enterprise contracts, not just early-stage startups.

The product is free right now, which is the standard land-and-expand playbook. Get teams using it, prove the value, then charge when they can’t live without it.

The Verdict

I think Oki is solving a genuine problem in a way that plays to AI’s actual strengths. Reading lots of scattered text and producing coherent summaries is something LLMs are legitimately good at. That’s a better foundation than most AI startups have.

The competitive risk is that this feature gets absorbed into existing tools. GitHub already shows activity summaries. Linear is getting smarter about project visibility. Jira could build this tomorrow (though knowing Atlassian, they’d take two years and ship something that requires a consultant to configure). The standalone play works if Oki can be the single pane of glass across all these tools, but that requires deep integrations that stay current as APIs change.

In 30 days, I want to see how many teams use Oki reports instead of their Monday standup, not alongside it. In 60 days, I want to know whether the stress monitoring feature is actually useful or just a novelty. In 90 days, the question is whether enterprise buyers will pay for this or whether it stays a nice-to-have that engineering managers love but can’t get budget for. The product is clean, the problem is real, and the team has the technical chops to pull it off. Whether “AI status reports” is a company or a feature is the billion-dollar question.