Search for a command to run...

Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
In this episode, Turner Novak sits down with Adit Abraham, co-founder and CEO of Reducto, a company that transforms PDFs and physical documents into structured data with human-level accuracy. (03:00) Reducto has processed over 1 billion pages, grew 6x in the past five months, and recently closed a $75 million Series B led by a16z while only burning $1 million in capital to get there. (00:42)
Adit Abraham is the co-founder and CEO of Reducto, a company that processes unstructured documents for AI applications. He has a finance background and learned sales through founder-led selling, taking the company to $5 million in ARR before hiring their first salesperson. Prior to Reducto, he competed in AI hackathons and initially worked on long-term memory for language models before pivoting to document processing.
Turner Novak is the founder of Gelt VC (formerly Banana Capital) and host of The Peel podcast. He focuses on early-stage venture capital investments and interviews founders about their startup journeys and scaling strategies.
Rather than trying to support 35+ file types like competitors, Reducto initially focused solely on PDFs and images - the 95% use case for most customers. (38:00) Adit explains how they would actually turn away customers if they couldn't solve their specific use case perfectly, preferring to build incrementally and ensure they were the absolute best at each capability before expanding scope. This approach led to much stronger customer satisfaction and word-of-mouth growth compared to competitors who offered broad but mediocre solutions.
Reducto landed a Fortune 10 customer as a two-person startup simply because their product worked when others failed on the customer's "gotcha documents." (32:00) The company built a playground where prospects could test the product immediately, letting the results speak for themselves rather than relying on sales tactics. This product-first approach enabled them to reach $5 million ARR with founder-led sales before hiring their first salesperson.
Contrary to traditional sales advice about not demoing on the first call, Adit found that technical founders have a unique advantage in being able to showcase product capabilities immediately. (65:46) He focuses on getting prospects to the "inspiration phase" - where they genuinely want to buy - as quickly as possible, often on the first call. This approach led to prospects bringing in C-suite executives on second calls because they were so excited about what they saw.
For complex challenges like chart extraction, Reducto didn't just train better models - they created an environment where AI could iteratively catch and correct its own mistakes. (21:37) The system renders charts, identifies errors, edits individual data points repeatedly, and uses a verifier model to ensure accuracy before delivering output. This systematic approach to hard problems demonstrates how to tackle challenges that can't be solved with a single-shot solution.
Reducto raised their Series B in just 48 hours because they built relationships with investors over time and only raised when they had genuine conviction about the value-add, not out of necessity. (56:00) They approached only two firms they were most excited about, ran a tight process, and focused on finding the right partner for their next growth phase rather than maximizing valuation. This approach worked because they had strong metrics and existing investor relationships.