Trendline Labs wants to close the information gap between casual and pro prediction market traders

Gregory Tolmochow graduated from Emory University last year with a degree in economics and computer science. Before he finished, he'd already co-founded a sports analytics company called The Betting Insider—aggregating unconventional stats like referee influence on NBA games, building it to around 250K followers, and turning it into what he describes as one of the go-to sports picking platforms in the space. He left in November, and a month later, he had a working MVP.

Tolmochow’s observation was that the same analytical infrastructure that existed for sports (ESPN, Action Network, a world of queryable data) didn't exist for prediction markets. Kalshi, Polymarket, Limitless, and now DraftKings and FanDuel were all spinning up their own markets, each with its own liquidity pool and its own pricing—but no one was building the intelligence on top.

That's what Trendline Labs is trying to be. Tolmochow describes it as a unified terminal for prediction markets—a platform where users can research, model, and trade across exchanges from a single interface. The product routes to the best available price across venues including Kalshi, Polymarket, Limitless, and PredictFun. But the more distinctive piece is Optimus, an AI model that gives context on why a market is priced the way it is—drawing on news, tweets, and Reddit threads to surface its own implied probability estimate. Users can also backtest models against historical data—something Tolmochow says Trendline has been building since day one, because no one else was storing it.

"Everybody was focused on being a unified brokerage," Tolmochow said on the podcast. "But now when somebody sends you a Bloomberg article, you trust it. And we want the same exact thing to be happening with Trendline." His read on the competitive landscape is that most rivals are optimizing for execution (latency, order routing, trading fees) for a high-frequency minority. He thinks that's a misguided direction. Ninety percent of Kalshi's revenue last year, by his estimate, came from the sports betting world: recreational users who want to understand a market, and not just trade it mechanically.

The regulatory path is less straightforward. Collecting trading fees in the U.S. requires an introducing broker (IB) license—a four-to-six month process that costs roughly $200K–$250K including legal and CFTC registration fees. Trendline is currently self-funded and seeking a $3M seed round to cover that process and get the product to market. For now, the platform is in private beta, with a full launch across prediction market exchanges expected in the coming weeks.

On the broader debate over whether retail traders can actually win, a question that's dominated prediction market discourse recently, Tolmochow pushes back on the shark-and-minnow framing. "Anyone can have an advantage with information," he said. "It doesn't need to be a pro." Unlike sportsbooks, where the house reprices continuously and bans winners, prediction markets let traders set limit orders, exit positions early, and buy low and sell high without ever holding to resolution. The information edge, his argument goes, is available to anyone paying attention.

That's the long-term pitch for Trendline: not another terminal for power users, but an intelligence platform that closes the gap between the casual trader and the pro.

Listen to the full podcast on YouTube, Spotify, and Apple Podcasts.