Radar makes podcasts searchable — and usable by AI agents

Particle, an AI newsreader startup founded by former Twitter engineers, is shifting its focus to a more lucrative idea: indexing spoken conversations within podcasts and making them discoverable. The company recently introduced Radar, a podcast search engine that transcribes audio and understands its meaning, enabling it to pull out key quotes and highlights.

Particle co-founder and CEO Sara Beykpour explained that hedge funds have been the highest-volume customers integrating with Radar's API, as they seek data inaccessible to their agents. Other top-paying customers include AI search platforms and data resellers. Radar's origins stem from a beloved feature of Particle's news-reading app, which sourced interesting podcast clips alongside related news stories.

The team saw the product's potential, but also recognized it as trapped within the news reader. With the rise of AI agents, Particle decided to pivot and focus on building an API for its podcast intelligence product. The company now transcribes over 130,000 podcasts, making it the largest transcribed podcast service available. These transcriptions include speaker labels, rich metadata, and speaker mentions across podcasts for tracking and alerting.

Users can customize alerts via email, Slack, or webhook, with filters for specific guests and topics. Radar can also extract self-contained clips, timestamps, and relevant metadata such as listener ratings, review, ad data, and political bias analysis. The API and MCP (Model Cloud Platform) allow businesses to programmatically tap into this intelligence.

Priced at $29 a month per seat, with a $399-per-month plan for businesses, Radar plans to expand beyond podcasts to support other forms of audio in the future.