Get to know: RotoBot AI

The co-founder of the "AI chatbot for fantasy football" reflects on the first NFL season behind him, and the opportunities ahead for RotoBot AI

Aishik Lala knew how he wanted to use his pre-med degree before even graduating from NYU.

He didn’t want to use it at all. So much (or little) so that by his sophomore year he’d begun working toward a minor in business technology management.

“I was like, ‘You know what? I'm already halfway through this. I'm not going to pivot so I can do another year of college’,” Lala recalled in a conversation with Betting Startups. “‘Let me just get straight B's and then get this degree.’”

The minor quickly proved both more stimulating and more immediately career-defining. At 21, he helped devise an app that the Miami Dolphins used to spark in-game fan engagement. Five years later, in October of 2024, RotoBot AI launched.

A large language model bolstered by data acquired through an agreement with FTN Data - founder Kevin Adams is an advisor and investor - RotoBot AI functions like ChatGPT but specializes in NFL fantasy football queries. 

“We saw a huge opportunity with the use case of Open AI and ChatGPT,” Lala explained. “We felt that some technology doesn't just come and go [and] typically, ecosystems are sort of built around this.

“Specifically for fantasy, we noticed there were tons and tons of people asking their sit/start questions all over the web, on X, on Instagram, on Reddit threads, TikTok. And really, every analyst's DMs are just flooded with these sit-start questions. 

“So we figured, why don't we just make an AI chatbot to go solve for this?”

RotoBot AI is currently available in the Apple Store, and will eventually migrate to Google, Lala said. After registering for the subscription-based service, needy fantasy players can interact through a conversational process that will seem immediately familiar.

It can be as simple as ‘Should I play Kelse or LaPorta?”

Or it could include ‘Why?’

Drawing on the paywalled trove of FTN stats and content, and scraping the Internet for complementary news about Kelse and LaPorta, RotoBot AI would offer a reasoned response in about five seconds. OpenAI still has a role: helping synthesize this wad of vector and SQL database information into a sensible sentence.

“We actually are not making just a chatbot. We’re currently dropping a news feed and different sit/start tools and different interfaces where it's not just a typing chat,” Lala added. “So that's also another thing that we want to bring. I would assume that ChatGPT is not going to have a fantasy football or NFL interface for at least the near future.”

The site, which launched last October, is currently in offseason maintenance mode, but could eventually expand into other sports leagues besides the NFL, Lala said. Or sports betting.

“If we see 40% of inputs are people trying to research their bets, that's a great indication for us on what we should do next. I think it's kind of a wait-and-see thing for us,” he said.

The pace of artificial intelligence development pushes the Massachusetts-based startup with 20 remote employees. Besides the core team of five full-timers, RotoBot AI boasts a group of about four content creators, three-to-four data scientists and five off-shore developers. 

“We're realizing that things that we thought would take years actually could take months instead,” Lala said. “The very first version of RotoBot that we released, which wasn't an app, it was just a web platform web app, that's something that could be built in a week.

“We took a handful of months to go build it, but someone can go build that in a week right now. We always want to maintain a really nimble mindset. We always want to do things the most efficient, quickest way because there's new tech dropping every day that makes everything that we've been doing even quicker. It's really exciting, kind of nerve-wracking, and confusing.”

Co-founder Aishik Lala (right) with the RotoBot AI team at The Starties in NYC

RotoBot AI is the second start-up for both Lala and chief technology officer David Zinland. While still in college, they built FlashPlays Live, a fan engagement app that allowed fans to make real-time predictions for prizes during sporting events. 

“The goal was to get into real-money gaming where people would actually bet on how many questions they would get right out of eight,” Lala, 26, explained. “Unfortunately, we saw too many legal hurdles. We were going to be available in only two or three states where people could actually bet.

“We had some problems there. We decided to flip B2B. We're like, ‘All right, let's create engagement apps for teams.’”

An accelerator program fostered a partnership with the Dolphins in 2021, and at one point the Dolphins Game Time app was the largest free-to-play, live sports prediction app on the market. 

With legal sports betting sapping the allure of free-to-play games, Lala and Zinland found full-time jobs. They reunited to devise RotoBot AI, adding Lala’s neighbor and family friend Ayan Chowdhury - the current head of data - and eventually head of experience Julian Mastroianni to form the core. Joe Hefley emerged from an iGaming and finance background to become the company’s primary angel investor.

Lala refers to this group as “techno optimists.”

“We believe in the future of technology,” he explained. “We believe in building the future of technology. And more importantly, we're a group of sports fans. We believe in the future of sports. We want to play a part in the innovation of it moving forward.”

It figures to happen rapidly.