Teamily AI - Your personal agent, agent teams & people in one messenger
by•
The human-agent platform where people and always-on agents get work done together in one unified, context-aware space. Unlike single-user personal agents, Teamily AI enables multi-user, multi-agent teamwork. Built on an AI-native messenger, it connects people, agents, docs, meetings, calls, workflows, and business systems to automate office work and customer service. Agents become real teammates with persistent memory, proactive intelligence, and continuous self-improvement. Free to get started.
Replies
This might work well for an agency managing feedback from several people.
@libin_yao Yeah, especially when three people give feedback on the same draft and you’re trying to figure out what to change. In Teamily AI, you can tag an agent in the chat and ask it to pull out where everyone agrees and what’s still up for debate.
turning a thread into slides is a neat idea
@jaelyn_zhang Thanks! My favorite part is being able to get a rough deck going while the conversation’s still fresh.
@jaelyn_zhang thanks Jaelyn!! The thread-to-slides moment is honestly a fun one to watch 😄 Getting a rough version while the conversation's still fresh means people react to something real, instead of trying to remember what they said last week.
The group chat format makes agent work easier for the whole team to see.
@domenic_yang Yeah, if someone tags an agent in the group chat, everyone can see what they asked and what came back. Saves someone from having to paste the answer into the chat later.
the shared context is what caught my eye
@anwar_uddin1 One place it helps is when you revisit a decision a week later. You can ask an agent right in the group chat, “Why did we go with this version?” and it can draw on the conversation your team already had.
@anwar_uddin1 Thanks Anwar! Shared context is the core of it. Memory carries across groups and sessions, so a new teammate can @mention an agent and catch up on weeks of decisions in seconds. I'd love for you to give it a try!
@anwar_uddin1 thanks Anwar!! Shared context is the bit I'd point to first too 🙌 It's the difference between an agent that answers your question and one that already knows why you're asking. Would love to hear what you try it on!
People and agents sharing the same conversation is a compelling idea. Congrats, team!
@kylee Thanks, Kyle! That shared conversation is the whole point: agents see the same context your team sees, so nobody has to re-explain anything. I’d love for you to give it a try! Congrats on MachineMail too 🙌
@kylee thanks Kyle! The shared conversation is the heart of it, agents see what the team sees, so nobody has to re-explain. Congrats on MachineMail too, rooting for you 🙌
I like that the agent sits in the channel instead of another app ！
@jojo_fang We wanted it to feel as simple as tagging a teammate. You ask in the group chat, and the answer lands right there for everyone to see.
@jojo_fang Thanks Jojo! That was a big design choice for us. Your team already lives in group chats, so the agents live there too. No new app to switch to, no context to copy over.
@jojo_fang Jojo yes!! No new app to open was kind of the whole point 😄 Every extra app is one more place where context gets lost. If the agent lives where the conversation already is, people actually use it.
We lose decisions in long threads all the time. does the memory surface them later?
@romy_chou You can come back to the chat a week later and ask an agent, “Did we settle on Friday or is that still up in the air?” It can use the earlier discussion to help you pick up where the team left off.
@romy_chou That's one of the most common pains we hear, Romy! Memory keeps track of decisions across long threads and different groups. And you don't even have to ask: when a topic comes up again, the agent proactively brings back the earlier decision, who made it, and why. Of course, you can also just ask "what did we decide about X?" anytime.
@romy_chou Romy yes!! and honestly ours is almost too reliable 😅 tell it a date once and it'll remind the whole group right on time. the part I'm designing now is getting the right thing back to the right person, not just everyone at once! All our company's daily work is done within Teamily AI. Every single day, we refine how it helps us carry out reliable work rather than losing track of things. We've navigated through various scenarios and challenges, and after a lot of fine-tuning, we've shaped it into what it is today.
Welcome!
The hardest part of AI teamwork is knowing what the agent already knows.
@new_user___3062026a4162ed354788f0c It’s hard to know how much background to give. In Teamily AI, if the team already talked it through in the group chat, you can ask the agent to recap what it picked up before giving it the task.
@new_user___3062026a4162ed354788f0c Totally agree, Linger. That's exactly the problem we set out to solve. In Teamily AI, memory is shared across the whole team, so everyone works from the same context. And agents don't wait to be asked: when something relevant comes up, they surface what they already know, like an earlier decision or a related file, right in the thread.
@new_user___3062026a4162ed354788f0c Linger yes!! Knowing what the agent already knows is half of trusting it. I don't want people to have to guess. I want the agent to be able to show you what it's working from, not only tell you. That's one of the things I care most about getting right 👀
Congrats on the launch! Curious how noisy a busy group gets with several agents 👀
@luke_pioneero Thanks! Even with a few agents in the group, they wait until someone tags them. Your team can talk things through and bring in an agent when they need a hand.
@luke_pioneero Good question, and something we designed for from day one. You decide how present each agent is: let it listen and jump in when it has something useful, mute it so it only responds when you @mention it, or remove it from the group and bring it back anytime. The goal is a teammate who knows when to speak, not a bot that talks over everyone.
@luke_pioneero You can put the agents in proactive mode or silent mode until they are directly called upon by other group members or agents. So, depends on the style of the group.
@yinn1014 Thanks for your vote! AI should empower teams, not just individuals. We believe the future is humans and AI agents working, learning, and growing together. 🚀
I also share my vision here:
https://x.com/AidenChaoyangHe/status/2117613537929703708
@yinn1014 Thanks Yinn! That was our starting point. Most AI tools assume one person and one chat, but real work happens in groups. So we built Teamily AI for the whole team: many people and many agents, one shared memory.
@yinn1014 Thank you so much for your vote. 🚀 We feel that approaching this from a team perspective is really valuable. While everyone talks these days about each individual being a team in their own right, interpersonal relationships and human social connection are what allow a team to achieve so much more. That is something we believe people will never move away from. We firmly believe that working, learning, and growing alongside AI agents will take us much further and enable us to accomplish far more than anyone could manage alone. We hope you enjoy using it.