Why We Invested in Stream
At TTV, we’re often asked about trends – which ones we’re following and where we’re allocating investment – and our answer is always the same: we don’t invest based on what’s popular or trendy. We invest in durable, sustainable businesses. We invest in hard-working, visionary entrepreneurs who are solving unmet market needs. It’s the same investment strategy that we’ve applied over the past 24 years, and it still rings true today when it comes to AI.
For us, it’s not about investing in AI because it’s everywhere and everyone else is doing it. It’s about making the right bets on AI. Some use cases are better than others; some market needs are proven, while others aren’t.
One example of a strong business case for AI is Stream, a real-time document processing platform that helps clients process workers’ compensation claims more quickly and efficiently. Stream uses AI to summarize documents, extract insights, and delete blank or redundant information, reducing the time it takes to process a claim from 7-10 days down to 1-3 days.
Document summarization is one task where AI is exceptionally skilled. LLMs have been trained on reading medical documents, and Stream itself has trained around 20 specialized models that are specific to workers’ compensation. The company’s AI is tailored to handle the use cases, reporting formats, and document types that claims adjusters manage each day. Workers’ compensation is a $50B industry, large enough to warrant a tailored solution. Stream has the numbers to prove there is high demand: in the past two years, their AI has reviewed over one million documents from clients. All of this serves as training data, allowing Stream to carve out a strong business case in the vertical.
Additionally, Stream’s co-founders, Eric Yen and Eilam Levitov, had direct experience working in the insurance industry. The two met while working at Next Insurance, where Eric was a product manager for the workers’ compensation line and Eilam was a software engineer. Eric started his career as a workers’ comp claims adjuster, where he saw firsthand that medical records are massive and time-consuming to review. Claims adjusters spend nearly 80% of their work day on document ingestion, reading, typing summaries, reviewing every claim, and making decisions throughout the process. The process was entirely manual, leading to long claim durations and errors. The workers’ comp industry is also behind when it comes to technology; at Eric’s first job, he was still using software that was built decades prior.
For any workers’ compensation claim, doctors need to review medical records, and they charge by the hour to review paperwork. Having blank or duplicate papers in a file adds to the expense and time needed. Further, the process lacks digitization – medical records for workers’ comp claims are primarily transferred via mail and fax. Each year, 10 billion pages of paper change hands between insurance companies and doctors so that an injured worker can get the medical care and equipment they need.
All of these factors are what led us to make our initial investment in Stream’s pre-seed funding in 2022, and to continue our investment by participating in the company’s recently announced $5.3M seed round led by Spark Capital with additional participation from Acrew Capital and angel investors Sam Hodges, co-founder and CEO of Vouch Insurance, and Mike Rosengarten, former co-founder and CTO of OpenGov.
With this new funding, Stream plans to further enhance its AI-driven platform, making it more powerful and adaptable to the evolving needs of the workers’ compensation industry. We are committed to supporting Eric, Eilam, and the entire team at Stream as they help clients process claims more efficiently so that people can get the medical care and equipment they need even faster. Stream’s AI is a model use case for the technology, and we’re looking forward to seeing how the company will expand its impact in the coming years.
