Serafis offers an API that indexes over 100,000 curated transcripts from podcasts, interviews, and independent media, providing semantic search and structured entity tags for companies, themes, and topics. The platform returns timestamped, source‑attributed quotes with relevance scores and integrates with data warehouses like Snowflake and BigQuery, enabling investment analysts and research teams to incorporate narrative intelligence directly into their analytics workflows.
Funding
$500K raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Investment professionals and data teams struggle to access timely, searchable insights from the growing volume of podcasts, interviews, and independent media where industry leaders discuss market trends. Traditional sources like SEC filings and earnings calls miss this unstructured narrative content, creating a research gap.
Solution
Serafis provides an API that indexes over 100,000 curated transcripts from podcasts, independent media, and other narrative sources. The platform enables semantic search by meaning, returning relevant segments with timestamped citations and relevance scores for companies and themes. Structured tags allow users to filter results by entities such as tickers or topics, while source-attributed AI queries ensure traceable quotes. The service can be integrated directly into research agents, data platforms, or analytics warehouses (e.g., Snowflake, BigQuery), delivering narrative intelligence in the tools analysts already use.
Target Audience
Primary customers are investment analysts, fund managers, and corporate research teams that need rapid, searchable access to narrative insights, as well as data engineering groups building AI‑driven research pipelines.
Features
- Semantic search across 100K+ curated transcripts, returning meaning‑based results rather than keyword matches
- Structured entity tags with relevance scores for companies, themes, and other concepts
- Source‑attributed responses that include exact quotes, timestamps, and media type (podcast, interview, etc.)
- API and MCP server for no‑code integration with LLM agents such as Claude, enabling automated research flows
- Direct data‑warehouse connectors for Snowflake and BigQuery, allowing SQL queries and joins with internal datasets
- Continuous index expansion with automated ingestion of new episodes and independent media content