Rob Williams

Rob Williams is the co-founder and CTO of Read AI. Many people know it as a meeting notetaker, but as he frames it, Read AI captures all of the conversations people have throughout their work days so it can do valuable things with them. He joined us in Episode 54 (October 9, 2026) and demoed a feature that had not shipped yet.

What Read AI does

Rob said meetings are, for many companies, the biggest investment of time, energy and cost they make (02:23). Read AI's goal is to get more out of them by collecting the scattered conversations inside a company and using them to prepare you for your day.

Read AI has 5 million monthly active users, Rob said, so the team paces its rollouts. The company started five years ago.

The demo

• The daily briefing (03:09). Rob showed what he woke up to that morning: a briefing built from email interactions and what Read AI knows about the podcast. It told him to prepare a demo for us. It was staging data, going to production the following week.
• Feeding it context (03:33). The briefing is not meant to be passive. Users can tell Read AI how to structure data and what is related.
• Folders, actions and rules (04:10). Launched over the summer, the folders view routes meeting data by rule. A client call can flow one way, while product discovery or user interviews go to a completely different group of internal users.

Key ideas from the conversation

• The adoption divide (05:18). Inside one company, some people think everything should be shared with no secrets, and others are terrified the tools will steal everything they know. Rob called handling that one of the bigger product and go-to-market challenges.
• Seven-day retention (05:53). Some organizations set retention to seven days. They treated older data as stale and more likely to mislead, so they shared fast and then retired it.
• Assistants move up the stack (09:16). Jordan mentioned a client whose executive assistants used to sit in on calls as live scribes. Rob sees assistants becoming air traffic controllers, deciding what information needs to travel through the organization.
• The information worker is the niche (10:31). Instead of one industry, Read AI focused on people whose work happens in meetings. An early use case was time shifting context between offices, like oil fields in Africa working with offices in Asia.

Every day that goes by that you're not capturing the conversations that are happening within your organization is a day that those conversations are gone forever.

— Rob Williams, Episode 54

I would have thought 30, 60, 90 would be the minimum, but there are companies, especially in some of the AI forward spaces, are just at the speed of light.

— Rob Williams, Episode 54

Building for models that keep improving

Asked how faster, cheaper models change Read AI (13:03), Rob said shrink-wrapped software of thirty years ago was a static noun. Teams now need to build a verb.

Like the organization should be building the crank that we can turn to update the entire product line to take advantage of new abilities quickly as possible.

— Rob Williams, Episode 54

In 2022 and 2023, Read AI chose not to pour energy into prompt engineering, assuming much of that work would be wasted. Instead, it built the rest of the product so model improvements would add value for free.

Next, Rob wants information to move faster (14:35). He walked through everything that has to be true for Sam to get an answer to a Slack message: Rob awake, at his computer, with the DM in front of him and a reply ready. He thinks Read AI can answer most of them already. The hard part is trust.

But I think in three years, let's say, which is a really long time, companies that don't allow their information to just move itself effectively to where it needs to be, to what's reasonable for others in the organization to know, they're going to be strategic disadvantage.

— Rob Williams, Episode 54

In the news segment (16:17), we covered Hot Girl Hotline, an AI relationship-advice app. Rob called it phenomenal marketing and branding (he once used Claude to unclog his disposal) and praised its unit economics. For more on AI at work, see our AI in productivity page.

FAQ

Who is Rob Williams?

Rob Williams is the co-founder and CTO of Read AI. He joined Built This Week in Episode 54 to demo the Read AI daily briefing and folder routing rules.

Is Read AI just a meeting notetaker?

Rob said many people know Read AI as a meeting notetaker, but he frames it as capturing all the conversations people have throughout their work days. What he gets excited about is what comes next, like daily briefings and routing context to the right people.

How many users does Read AI have?

Rob said Read AI has 5 million monthly active users, which is why the team paces rollouts of new features like the daily briefing.

Why would a company set Read AI data retention to 7 days?

Some fast-moving, AI-forward companies treat data older than seven days as stale and more likely to lead people to be wrong. They share it far and wide, share it fast, then retire it.

How does Read AI plan for better AI models?

Rob said the team chose not to over-invest in prompt engineering in 2022 and 2023. Instead they build the rest of the product so expected model improvements add value on their own, which he calls building a verb rather than a static noun.