Launching today
Semos.ai Manager Agents
AI agents purpose-built for managers
398 followers
AI agents purpose-built for managers
398 followers
Manager Agents learn from your meetings, then point you to what needs your attention next: the feedback that is overdue, the recognition you missed, the conflict you are avoiding, the growth talk that keeps slipping. You learn what works, what doesn't, and why, so you become better at leading your people.
Semos.ai Manager Agents
Hey Product Hunt 👋
Our team has spent years building people and culture intelligence software for large enterprises at Semos Cloud. Working closely with those companies, we saw a gap in the tools available to help managers lead their teams. We wanted to give them something practical they could use in their everyday work. That became Manager Agents.
Manager Agents learns from your meetings and helps with the parts of managing people that usually get pushed to the bottom of the list: feedback that’s overdue, recognition you almost missed, the conversation you’ve been avoiding, the growth talk that keeps slipping.
This isn't just another AI chatbot. Specialized agents handle different parts of leading a team, all working from the same context pulled from your meetings, so the guidance stays grounded in your actual team rather than a generic answer. We also draw on behavioral science to shape the guidance, including how to give feedback and approach conversations with different people.
A few things it does:
💬 Drafts feedback and recognition messages grounded in what actually happened in your meetings
🎯 Helps you prep for the hard conversations before you have them
📊 Gives you a real read on how you're leading, not just a guess
🔔 Nudges you when something's worth acting on, so it doesn't fall through the cracks
Who it's for: people managers with a handful of direct reports, doing their own work at the same time, often without an HRBP down the hall or any formal training. If that's you, this was built with you in mind.
Try it and let us know how it goes and where it could be better 🙏
Semos.ai Manager Agents
@natasha_davidovska So proud to be part of this team! The "without an HRBP down the hall" part is so real. That's most managers I know, including me. Can't wait to hear what everyone thinks 🚀
@natasha_davidovska Does it track patterns across multiple meetings over time ?
Can the agents learn from that kind of correction?
Semos.ai Manager Agents
@lanfranco_iwanaga Hi Lanfranco, yes, that's actually one of the core ideas behind how Manager Agent work. The agents build context from your meetings over time, so the more you use them, the more specific the guidance gets to your team, your style, and how you lead. If you correct something or push back on a suggestion, that becomes part of the picture too, it's not starting from zero each time. Thank you for your question!
on device processing or strict data encryption for sensitive leadership meetings is crucial here.
Semos.ai Manager Agents
@sandy_leventh Hey Sandy! Completely agree, and that's something we take seriously. Personal information is masked with PII masking before anything is passed to the AI models. On top of that, data is encrypted in transit, so meeting content is protected at every stage, not just when it's being processed. Thank you for your question, we appreciate it! 🙌
I think remembering to give feedback is harder than people admit. Does it also suggest what kind of feedback to give?
Semos.ai Manager Agents
@luke__nick Totally agree, Luke! And yes, it does. It drafts feedback based on what actually happened in your meetings, so it's specific rather than generic. It also draws on behavioral science to help you tailor how you say it to each person. You stay in control of the final message, it just gives you a strong starting point, based on Stanford’s 4Is framework. You can read more about the science behind here https://semos.ai/explore-agents-new/
How does the system distinguish between a real unresolved issue and something that was simply mentioned during a meeting?
Semos.ai Manager Agents
@mariusholm Hey Marius! 😊
Every signal goes through a few stages before it ever reaches the manager. It starts as detected, then gets verified against multiple meetings or sources, and only after that does it get published as something you'd actually see.
So if something's just mentioned once in passing, it stays at that first "detected" stage and won't surface as a signal on its own. It only becomes a real, validated signal once it's come up more than once, ideally from more than one person too, which is exactly how we tell a passing comment apart from something that's a genuine, recurring issue.
Does it help managers look back at previous conversations and see whether certain approaches led to better outcomes?
Semos.ai Manager Agents
@bastienrenard Hey Bastien, thanks for the question!
Not in the sense of running a "this approach worked better" analysis, but it does something close to that. Manager Agents builds context from meetings over time, so it can show you things like a shift in tone across your 1:1s with someone, or how recognition and feedback have looked over the past few weeks or months, not just one conversation on its own.
So if you gave a certain kind of feedback and things changed with that person afterward, that's the kind of thing it will show you. It's less "Approach A beat Approach B" and more "here's what's changed since you handled it this way."
There's also a Prep for a Meeting feature, which helps you walk into a meeting better prepared using exactly this kind of context, past patterns, what's come up before, so you're more likely to get the outcome you're looking for.
I would be curious to see how much control managers have over what gets surfaced.
Semos.ai Manager Agents
@henryevans11 Hi Henry!
Right now, you see everything the agents pick up on (feedback and recognition gaps, potential conflict, growth signals, etc), there's no way to say "only show me feedback-related things" or mute a category you don't want to see.
You can choose what to do with the information, if you want to act on it or not, it won't take any action on your part.
Thanks for the question, it's a good direction for us to explore.