Podcast guest program
Synthetic-voice podcast segments
The Data Nerd records 5, 7, and 10-minute podcast segments via synthetic voice, the same voice that powers our methodology walkthrough. Hosts receive a verbatim 60-second opening, five Q&A prompts, and links to receipts (SSRN paper, methodology, dataset). The format is methodology-first; the founder identity is incidental.
Prepared topics
Seven segments, ready to ship. Each comes with a verbatim opening transcript, host-Q&A questions, and a receipts pack for show notes.
7-min segment
GitHub Commit Velocity Is the Most Leading Public Signal in VC. Here's the Panel.
How a 219-company longitudinal panel showed a median 21-to-47-day lead time over Series A announcements.
- · AI/ML researchers curious about applied alternative-data work
- · Venture capitalists wondering if public data can produce real edge
- · Engineers who want to know what their own commit graph signals to outsiders
7-min segment
Code-Side Sourcing: Naming the Category That Sits Between Crunchbase Scraping and Warm Intros
Why the third era of VC sourcing, reading public engineering activity at scale, needed a name before it could be priced.
- · Emerging-manager GPs building thesis-led funds
- · VC platform teams evaluating which alternative-data layer to buy
- · Founders curious how investors will discover them in the next 24 months
7-min segment
What Agent-Native VC Tooling Actually Looks Like, and Why We Charge Per Call
Free MCP server, EUR 0.19 per deep-signal call via x402 USDC, no subscription gate. Here's why.
- · AI engineers building investor-facing agents
- · VCs experimenting with Claude/Cursor for diligence workflows
- · Anyone curious how x402 micropayments will reshape API pricing
5-min segment
Why a Venture-Capital Product Published Its Methodology on SSRN, And Re-Released the Data CC BY 4.0
Open data is not anti-business; it's the only defensible moat when the next product is shipped by a team three times your size.
- · Open-source maintainers wondering if 'open core' works in VC tooling
- · Researchers curious whether industry data ever makes it into peer review
- · Founders evaluating moat strategies in commoditizing markets
10-min segment
How to Read 350+ GitHub Organizations a Week Without Getting Rate-Limited Into Oblivion
The systems story behind the panel, five primitives, three layers of caching, one war with the secondary-rate-limit ledger.
- · Senior engineers doing data-pipeline work at scale
- · SREs curious about the GitHub API rate-limit topology
- · Anyone building public-LLM-grounded products at scale
5-min segment
The 38 Percent False-Positive Rate We Won't Hide, And Why That's the Whole Point
Why publishing the failure rate of your signal is the only credible move, and how 'silent rounds' broke our intuition.
- · Data scientists who love a good post-mortem
- · Investors who want a rigorous take on alt-data overclaiming
- · Anyone who's tired of vendor decks claiming '95% accuracy' with no error bars
7-min segment
We Gave Away the MCP Server. Here's What It Did to the Top of the Funnel.
How a free six-tool MCP server became the highest-converting acquisition channel for a paid VC product.
- · Founders deciding whether to launch an MCP integration
- · Indie hackers exploring agent-native distribution
- · AI/ML engineers shipping their own MCP servers and wondering if the funnel works
Active outreach
15 shows in our ready queue. Updated as shows respond.
Hosting a show that fits? Email press@gitdealflow.com with the show URL and audience demographic. We’ll send the verbatim 60-second preview and a synthetic-voice sample within 48 hours.