Particle Launches Radar, a Podcast Search Engine and API for AI Agents
Particle, the AI newsreader startup founded by former Twitter engineers, launched Radar on Wednesday - a podcast search engine that transcribes and semantically indexes more than 130,000 shows and exposes that data through an API for AI agents and enterprise customers.
👁 116 reads
Particle, the AI newsreader startup founded by former Twitter engineers, launched Radar on Wednesday - a podcast search engine that transcribes and semantically indexes more than 130,000 shows and exposes the results through an API targeting AI agents, hedge funds, and data resellers.
- Radar transcribes more than 130,000 podcasts and adds 20,000 episodes to its index daily, making it the largest transcribed podcast service according to Particle
- Hedge funds are the highest-volume API customers, with AI search platforms and data resellers also among top paying users
- The API surfaces speaker-labelled transcripts, entity metadata, timestamped clips, and ad tracking - giving AI agents access to audio that they would otherwise be blind to
The product transcribes every episode with speaker labels and rich entity metadata, recognising the people, companies, brands, products, and topics under discussion. According to the company, 20,000 new episodes are added to the index daily, covering all Apple Top 200 podcasts across 135 verticals. Particle describes Radar as the largest transcribed podcast service in existence.
Co-founder and CEO Sara Beykpour told TechCrunch that financial firms have emerged as the most active integrators of the API:
“Hedge funds have been the highest-volume customers that are directly integrating with the API.”
AI search platform Exa is among the named partners already using the data layer. The rationale for the product is straightforward: most web-crawling agents operate entirely on text. As Beykpour put it, “Agents are generally blind to audio; they can’t see it unless something or someone has transcribed it.”
Beyond raw transcripts, Radar can track entity mentions across the podcast catalogue and dispatch alerts via email, Slack, or webhook - on a real-time, daily, or weekly basis - with filters that let users trigger notifications only when a specific guest discusses a particular topic. The system also extracts self-contained timestamped clips rated by Particle as the most notable moments in each episode. A dedicated podcast-ad search engine is included as well.
The pivot grew out of a feature inside Particle’s original news-reading app, which sourced podcast clips to sit alongside related news stories. The team recognised the data layer had broader value but was constrained inside a consumer product - and the rise of agentic AI workflows gave it a commercial reason to break it out as a standalone API.
Practitioner takeaway: Developers building research, monitoring, or intelligence agents now have a structured audio data source to query rather than scraping raw RSS feeds or relying on incomplete third-party transcripts. For small teams doing competitive tracking or media monitoring, Radar’s webhook-based alerts and entity filters offer a way to pipe podcast intelligence directly into existing workflows without building a transcription pipeline from scratch.
Google · Your sourcesQuantslant higher in your Google - one clickAdd to preferred sources →
More on this from QUANTSLANT
news·Z.ai Confirmed as Creator of Benchmark-Topping Mystery Model Ox Alpha