Episode 141: From Notes to Action: Fellow AI’s Vision for Workflows

We are getting to a place where AI will become proactive, like Chat GPT's pulse which uses your activity overnight to do research and provide relevant information based on your interests the next morning.

Hi, my name is Demetri Panici, and I'm a content creator, agency owner, and AI enthusiast. You're listening to the AI Agents podcast, brought to you by Jotform and featuring our CEO and founder, Aytekin Tank. This is the show where artificial intelligence meets innovation, productivity, and the tools shaping the future of work. Enjoy the show.

Hello. Welcome back to another episode of the AI Agents podcast. In this episode, we have Aiden Mirzaie, the CEO and founder of Fellow AI. How are you doing today, Aiden?

Great. Thanks for having me on, Demetri.

I love what you're doing. AI note-taking is one of my passions with all the different tools out there. Outside of this podcast, my YouTube channel got me into content for AI and SaaS companies, which sparked my interest in tech and AI in general. I'm excited to talk with someone in my realm who makes you more productive with AI-powered tools.

First and foremost, tell us a little bit about how you got into the space of note-taking and how you went from where you started to implementing all this cool AI stuff at a meeting company.

Great question. I have an engineering background but have always been an entrepreneur. I've started three companies with my co-founders and five different products. Our last company was in the online survey space with a product called FluidSurveys, which we sold to SurveyMonkey. I was at SurveyMonkey for a few years leading up to the IPO, then my co-founders and I left to start Fellow.

The original idea behind Fellow was to build a tool for managers, but we realized to help managers, we had to tackle meetings because they spend all their time in meetings. That became the highest surface area. The first part of Fellow's journey was building a tool to make meetings effective and help people spend less time in them.

As COVID happened and more companies became remote, meetings exploded in number. Then the Chat GPT moment happened at the end of 2022, and we saw meetings would be revolutionized. AI's best application would be in meetings. We reinvented our product and company to deliver the same value using AI, which was a big transformation.

Over the last two years since pivoting, we went from zero to eight figures in AI revenue. The company is in the best place it's ever been, fastest growing, in a more mature but rapidly changing market. We talk about where we're going and the niche we'll dominate. Everyone's heard of AI note-takers, but the question is how we're winning.

You managed to change your product quickly and make eight figures in AI revenue. I have a question about your product. Many people have experienced tools in Google Meet, Microsoft Teams, or Zoom that pop in with a bot name. Tell us what it's like to have Fellow AI inside a call from the perspective of people on the call who aren't using it and those who are.

Fellow is designed to be effortless with two modalities. One is Fellow joining as an attendee, showing up as another tile on your screen, useful if you want your agent to attend meetings on your behalf. The other is a bot-free modality where Fellow runs in the background taking notes without showing a tile. Both parties can see the call is recorded to build trust and avoid surveillance concerns.

Fellow supports all meeting types including in-person, Slack huddles, and phone calls. We believe no one should be recorded without knowing. Our approach is collaborative so everyone shares the same action items list, avoiding multiple note-takers and confusion. Privacy and security are core philosophies that differentiate Fellow.

During meetings, Fellow provides real-time feedback with a live chat powered by AI models like Chat GPT or Claude. It offers suggestions, counterpoints, and smart follow-up questions to help you be on top of your game. After meetings, it automates follow-up emails, CRM updates, knowledge base articles, and clips shared in Slack, acting as an always-on AI chief of staff.

How did you figure out what features people find awesome versus ones you didn't pursue? Customers often suggest features grounded by what they know is possible, but AI expands possibilities rapidly. You have to imagine compelling futures and not only build what customers ask for. Inspired by Brian Chesky's 10-star experience concept, we envision making every meeting a 10-star experience with AI helping prep and execution.

Meetings are constructs that help companies operate, but with AI, we question what meetings mean when multiple agents participate. We imagine futuristic organizations and bring elements back to reality. There are six and seven-star hotels in the world, showing experiences can go beyond five stars. Software and AI are about delivering the best experience possible to attract users.

On integrations, Fellow pushes action items to various task management systems like Asana, Monday, Google Tasks, Microsoft Tasks, and CRMs. It's hard to standardize because everyone prefers their own system, so we integrate with all. Getting action items right is critical because missing one loses trust. We use agents that argue to determine actual action items, avoiding flooding users with irrelevant tasks.

We also update CRM fields intelligently by running prompts on meeting discussions to fill complex fields like customer potential. This flexibility extends to wikis and other systems. Our goal is to be the agent that communicates with all tools to consume and distribute knowledge effectively.

We support MCPs so users can have a primary AI. For example, developers use Cursor with Fellow MCP to consume meetings and produce code based on discussions, short-circuiting workflows. We also have native connectors for Claude. No matter the AI used, connecting tools lets you consume knowledge in any way you want.

You can automate creating tickets in Linear or Jira from feature requests and bugs mentioned in calls. This automation makes every employee act like your best employee by removing friction and ensuring consistency. Proactiveness is key because people often delay tasks due to blockers, but AI can prevent or burst through that friction.

Today, AI interaction is mostly ask-based, but we're moving toward proactive AI that anticipates needs. Chat GPT's pulse is an example, researching overnight and delivering relevant info based on your interests. This proactive AI will be the next stage, changing how we interact with technology.

The world of MCPs and AI agents is opening up possibilities for everyone. Tools like Claude Code and sub-agents make advanced workflows accessible even to non-developers. As AI becomes more integrated, people will rely on it more for problem-solving and productivity.

Our instinct is shifting from relying solely on ourselves to consulting AI first, similar to how we use GPS for driving. GPS knows traffic and road conditions better than we do, so we trust it for the best route. Similarly, AI will provide context and solutions beyond our capacity, becoming a trusted assistant for many problems.

Some people worry about offloading thinking to AI, but it's more about outsourcing initial steps and having the option to analyze or override suggestions. Just like you might ignore GPS if you know of a traffic jam, AI is a tool to augment, not replace, human judgment.

Minimalism in choices, like wearing the same black shirt daily, frees mental energy for more valuable tasks. Similarly, using AI for routine decisions gives us more brain power for important thinking. The option to use AI or not is there, giving us more flexibility and efficiency.

Meetings used to primarily benefit participants, but now they produce context, the most important asset for AI. Capturing meeting context allows remixing content into memos, wiki articles, or AI problem-solving inputs, making every meeting benefit the entire organization and making it smarter over time.

This is the power of AI in meetings, turning them into organizational knowledge generators. Fellow is at the forefront of this direction, helping create modern, futuristic organizations where meetings and their meaning evolve. You can find Fellow at fellow.ai and check it out.

Thank you everyone for listening. If you liked this episode, hit like, subscribe on YouTube, and leave a review on Apple Podcast. We appreciate your support and will see you in the next one. Bye.