Radar Podcast Search Engine: Full-Text Search Across 130,000+ Shows with AI Agent Integration
Radar transcribes 130,000+ podcasts into a full-text index accessible to AI agents via API and MCP.
Radar, from the Particle team, is a podcast search engine that transcribes and indexes 130,000+ shows, filling a long-standing gap in audio searchability. Its dual value lies in giving users granular, content-level search across podcasts while exposing that knowledge base to AI agents through the Podcast Intelligence API and MCP protocol. The article also flags transcription accuracy and copyright compliance as ongoing challenges the product will need to navigate.
The Search Gap in the Podcast World
Text content on the internet has long been highly searchable — you can Google any blog post, news article, or forum thread. But the audio world remains a kind of "dark matter": countless podcast episodes are published every day, carrying high-quality, timely insights and conversations that are almost impossible to search effectively.
Radar, the latest product from the Particle team, is built to close exactly that gap. It positions itself as "The Podcast Search Engine," built around a simple but powerful idea: if web pages can be searched, podcasts should be too.
After launching on Product Hunt, the product earned 83 upvotes and ranked 12th for the day, placed across three categories: Developer Tools, Artificial Intelligence, and Tech.
Core Features of the Radar Podcast Search Engine
Large-Scale Podcast Transcription and Full-Text Indexing
According to the official description, Radar currently covers 130,000+ podcasts with ongoing transcription, adding roughly 20,000 new episodes per day. In aggregate, that means millions of hours of audio and billions of lines of text have been fully transcribed and indexed — all searchable.
That scale is significant for podcast search. Traditional podcast platforms like Apple Podcasts and Spotify rely primarily on titles, show notes, and tags. Users have almost no way to track down a specific thing a guest said in a specific episode. By converting audio to text and building a full-text index, Radar brings search granularity down to the level of individual spoken statements.
From "Find a Show" to "Find an Idea": A New Dimension of Podcast Search
This shift represents a fundamental upgrade in how podcast content can be explored. Previously, you either remembered a show's name or stumbled onto relevant content by chance. With a searchable text index, users can search directly for a topic, a person's name, or an event — and surface every relevant discussion across thousands of shows.
For researchers, journalists, and content creators, this effectively transforms podcasts from a passive entertainment medium into a citable knowledge base.
Podcast Intelligence API for AI Agents
API and MCP Protocol Support
Radar earns its Developer Tools label because it's not just a search box for end users. Behind it is Particle's Podcast Intelligence API, which supports AI agents in searching across podcasts directly via API and MCP (Model Context Protocol).
This is worth emphasizing on its own. As large language models and agent ecosystems mature, MCP is becoming an important standard for connecting AI to external data sources. By exposing its podcast index through MCP, Radar enables any connected AI assistant to treat global podcast content as a callable knowledge source.
The Agent-Ready Value of Audio Data
Most mainstream AI applications today draw from web text, while audio content has long been overlooked due to the difficulty of processing it. Podcasts often contain the freshest industry takes, expert opinions, and firsthand information — content that tends to be more immediate and candid than formally published articles.
Radar converts this high-value audio data into a structured interface that agents can consume, effectively giving AI systems a "sense of hearing" they've been missing. For developers building research assistants, market intelligence tools, or content summarization products, this opens up a data source that was previously very hard to access.
Use Cases and Current Limitations
Four Key Application Areas
Radar has clear value across several directions:
- Industry intelligence monitoring: Track how specific topics evolve in discussion across many podcasts over time
- Research and precise citation: Quickly locate an expert's original words on a given question
- Content creation and topic discovery: Help creators find story angles and reference material
- AI product integration: Serve as a real-time knowledge backend for agents, enriching AI systems with audio-sourced information
Challenges Worth Watching
That said, products like this face inherent challenges. First is transcription accuracy — automatic speech recognition can still make errors with technical jargon, multiple speakers, and accents, which directly affects search quality. Second is the copyright and compliance boundary: how large-scale transcription of third-party audio content balances fair use against rights holders' interests remains a grey area across the industry. And while 130,000 shows is impressive, it still represents a fraction of the total global podcast catalog, leaving considerable room for expansion.
Closing Thoughts: Audio Content Enters the "Programmable Knowledge" Era
Radar reflects a clear product instinct: the searchability of information shouldn't stop at text. In an era when AI agents are rising fast, structuring, indexing, and making programmable the vast, underexploited knowledge locked in podcasts serves both human users who need better search and AI systems hungry for high-quality data.
For developers and information workers, Radar is worth paying attention to not just for its podcast search functionality, but for what it represents — opening audio knowledge to the AI ecosystem through API and MCP. That may well mark the beginning of audio content's formal entry into the age of programmable knowledge.
Related articles
Andrew Ng's Agentic AI Course Distilled: Core Methodology for Building AI Agents
Andrew Ng's Agentic AI course decoded: cut through the hype, build real value with disciplined Evals and error analysis. Key insights for AI agent developers.
iRobot Roomba Duo Dual-Robot Concept: Exploring a New Form Factor for Robotic Vacuums
iRobot debuted the Roomba Duo concept at IFA — a dual-robot system pairing a heavy-duty floor washer with a slim Roomba to tackle hard-to-reach areas.
Confessions of a Heavy Gemini User: 3 Hours a Day, and How AI Dependence Erodes Independent Thinking
A Reddit user confesses to 3+ hours daily on Gemini, outsourcing everything from coding to life choices. We explore AI dependency, cognitive offloading, and how to protect independent thinking.