Startup Spotlight

Each week, we profile one early-stage startup from our submission pool — companies you won't find covered elsewhere. Some are raising their first rounds. Others are building in stealth, gathering early users before they go to market. This week: Ravioli.

Prediction markets have grown rapidly, with platforms like Polymarket and Kalshi generating over $44 billion in notional trading volume in 2025. But the tradable events on those platforms share a common limitation: they only resolve objective outcomes. Who wins the election? Who wins the Super Bowl? Left off the table is the category of debate that dominates online attention every day, questions without a verifiable answer. Are the ICE raids justified? Is remote work better for productivity? These subjective arguments play out across Reddit threads, comment sections, and social media feeds, generating enormous engagement with no mechanism for participants to put stakes behind their reasoning. The result is a lot of noise, with no effective way to surface which arguments actually hold up.

Ravioli is building a platform where users can trade on the outcomes of subjective debates. The mechanism that makes this possible is a proprietary logic scorer: an analytical system that evaluates the quality of reasoning in posts and comments, rather than simply fact-checking claims. When a debate is opened on the platform, the scorer assesses the strength of each side's arguments, and users can back whichever position they find more persuasive. The company describes this as going significantly further than conventional fact-checking by analyzing the structure and coherence of reasoning itself, not just whether individual claims are accurate. The rewards aspect of the platform currently operates under the Ravioli name, while a companion platform, Tilt, runs without the financial component as the team navigates the regulatory requirements needed to deploy rewards at scale.

Ravioli was founded by Wyatt Wolfman and Vir Sanghavi. Wolfman previously built and scaled a resale business to close to $250,000 in annual revenue while still in high school, then went on to found three additional companies, giving him early-stage operations experience across multiple ventures. Sanghavi is a former Texas State champion and nationally ranked congressional debater, which informs the core of what Ravioli does: scoring discourse without bias requires genuine expertise in how arguments are constructed and evaluated. On the technical side, Sanghavi has built ML-based facial recognition software and AI-driven bird strike prevention systems for aircraft, and placed third out of 200 participants at Y Combinator's Agent Jam hackathon.

Ravioli launched its Tilt social platform without paid marketing and grew it to more than 5,000 users. The company's Instagram presence has reached over 11,000 followers, with millions of views across social channels, also without paid spend. Both figures suggest early organic demand for the concept among the politically engaged, platform-native audience the company is targeting.

Prediction markets recorded over $44 billion in notional trading volume in 2025, according to data from Keyrock and Dune Analytics, with monthly volume growing from under $100 million in early 2024 to over $13 billion by year's end. Ravioli is not positioning itself directly against Polymarket or Kalshi, which focus on objective outcomes. The company's thesis is that subjective debates, which generate significantly more daily engagement online than questions about sports outcomes or election results, represent an adjacent and largely unaddressed opportunity within the broader event-trading category. The regulatory landscape for prediction markets continues to develop, which presents both a challenge and a timing consideration for the team.

Interested in learning more? Contact Wyatt Wolfman at [email protected] or visit ravioli.live.

Know a startup building something interesting? Submit for consideration: