AI companies are pushing agents as the next Great Workplace Disruptor, but experts say they’re still not ready for prime time. AI agents often struggle to make decisions by themselves, hallucinate frequently, can’t cooperate with other agents, fail at confidentiality awareness, and integrate poorly with existing systems.
Industry pioneers like Andrej Karpathy and Ali Ghodsi have said that, like the deployment of autonomous vehicles, humans need to be in the loop in order for agents to succeed.
A startup called Mixus wants to address that with its AI agent platform that not only keeps humans in the workflow, it also lets users interact with agents directly from their email or Slack.
“We’re meeting customers where they are today,” Elliot Katz, Mixus’ co-founder, told TechCrunch. “Where is every person in the workforce today? For the most part, they’re on email. And so because we can do this through email, we believe that’s a way we can democratize access [to agents].”
If Mixus works reliably, it may solve a big problem in the AI agent space. Most AI companies today either give you a pre-built assistant, à la ChatGPT or Gemini, or developers have to build custom agents using frameworks like LangChain, AutoGen, or crewAI.
Mixus launched in beta out of Stanford only in late 2024, but it has already raised $2.3 million in pre-seed funding and brought on some customers, including clothing store chain Rainbow Shops, as well as others across finance and tech.
The startup says its biggest selling point is ease of use, from how it helps you create agents to how you can interact with them. Users can use text prompts to set up their agents within Mixus’s platform via a chat function, or by simply emailing instructions to agent@mixus.com. Then Mixus will build, run, and manage single- or multi-step agents directly from the inbox.
Techcrunch event
San Francisco
|
October 27-29, 2025
For example, a salesperson may use a prompt that reads so:
Create an agent that finds all open tasks in Jira in project mixus-dummy, and send me a report with information on all tasks that are overdue. Draft emails to all the assignees who have overdue tasks, and have me review them in the chat and with simple clear formatting for email (no attachments/docs). Once I verify, send the emails. Run it now. And moving forward, run it every Monday at 7am PST.
Katz and his co-founder Shai Magzimof demoed the agents for TechCrunch, showing how to add human verifiers for your agents by simply instructing at which step the agent should ask you for oversight.
For example, they ran an agent to do research on TechCrunch reporters before pitching them. The agent identified and gathered technology news and trends, analyzed the information to identify potential story angles, and compiled a research report summarizing the findings. At the last stage, the agent was directed to send the information to Katz for verification. Once approved, the agent would send the completed research report to Magzimof.
The founders noted that humans can be in the loop as much or as little as required — Magzimof said organizations can set up company-wide rules, like ensuring an email gets checked by a human if it’s being sent outside the company.
Bringing other colleagues into the workflow is as easy as tagging them in the chat with an AI agent, or copying them on the email to the agent. That’s another standout compared to agents on the markets today: Most models are single-user, and while Notion AI and Slack allow users to collaborate in shared spaces, they don’t let the AI manage conversations and tasks between teammates in real time.
Another core feature of Mixus is its ability to remember files, chats, prompts and agents.
“We created Spaces so that every team, every person, every group of people can have a shared memory,” Magzimof said. “Then all my agents, all my files, all the people can be in that very specific Space’s memory.”
While ChatGPT and Claude both support memory, their enterprise plans don’t yet support shared agent memory across users.
What else can Mixus do?
In our interview, the founders ran through an hour-long demo showing a range of use cases and abilities. Mixus’ agents do seem capable, reflecting a high degree of autonomy and memory that places the company towards the higher end of the AI agent spectrum. That is, if the product works as reliably as it did in the demo.
Like other agents, Mixus can integrate with other tools, from Gmail to Jira, and users can trigger agents to run immediately or on a schedule. Agents can run and edit documents or spreadsheets inline — similar to ChatGPT, Microsoft Copilot, and Google Gemini, but those are often limited to sandboxed environments.
Mixus also lets agents autonomously navigate organizational context — like figuring out who in an organization owns a task by looking through Jira tickets.
Built on a combination of Anthropic’s Claude 4 and OpenAI’s o3, Mixus agents also have access to the web, which Magzimof says can be tapped for tasks like live research or monitoring. He described it as “Google Alerts on steroids.”
Taken together, Mixus appears to be less of a productivity tool and more like a tireless digital colleague – yet another ambitious attempt to reimagine AI as a collaborator. If it works as advertised, your next “coworker” might not be human, but it might get through your inbox faster than you do.
Got a sensitive tip or confidential documents? We’re reporting on the inner workings of the AI industry — from the companies shaping its future to the people impacted by their decisions. Reach out to Rebecca Bellan at rebecca.bellan@techcrunch.com and Maxwell Zeff at maxwell.zeff@techcrunch.com. For secure communication, you can contact us via Signal at @rebeccabellan.491 and @mzeff.88.
Read the full article here
