A Team of One — Anekha Sokhal, Moshi
How Moshi turns market data into actionable insights, and why AI tools must fit right into a trader’s workflow.
AIE CODE tickets and applications are now open — the final flagship AIE event of the year.
Back when Anekha Sokhal was trading energy derivatives, staying on top of the market meant parsing through vastly more information than any human could realistically read. Today, as CEO of Moshi, she is building the tool she wished she had back then: a recommendation engine that connects market noise directly to what actually matters for a trader’s book.
At AI Engineer New York, Anekha sat down with swyx to discuss her journey from studying mathematics to JPMorgan, and eventually founding Moshi. She also talked about why practical AI must integrate seamlessly into how people actually work. She describes Moshi as a blend of Bloomberg, Claude, and Netflix: market data paired with an interface to interact with it, complete with personalized recommendations tailored to individual traders.
They also dug into trusted data sources, document extraction, the build-versus-buy dilemma for models, and the mounting pressure on AI engineers who are told “you are a team of one.” Anekha also shared key takeaways from talking to customers at the conference and explained why speaking with users has to come before hiring.
Watch the Episode
Watch Anekha Sokhal’s interview on YouTube
Show Notes
Moshi — Anekha’s company
Anekha Sokhal — Background and resume
GIQ Research Paper — The research discussed in the interview
Reducto — Document extraction tool
Timestamps
0:00 — AI Engineer New York & “a team of one”
0:54 — Getting to know Anekha Sokhal
2:09 — From math to energy derivatives
3:46 — Winning a Fulbright scholarship and becoming a founder
5:45 — What Moshi does for traders
6:10 — Managing the flood of market data
9:03 — Bloomberg and market workflows
10:03 — The intersection of Bloomberg, Claude, and Netflix
12:03 — From dashboards to workflows
15:03 — Trusted sources and actionable signals
17:13 — Build vs. buy for models, and document extraction
18:20 — On the ground at AI Engineer New York
20:32 — Customer meetings and community conversations
23:54 — Hiring and customer discovery
25:55 — swyx’s path in AI
Transcript
Automatically generated transcript; phrasing may contain minor errors. Speaker labels omitted.
0:00 I actually ended up extending my stay in New York because right now—I was supposed to leave today. But then meetings lined up for Thursday, so I stayed an extra day; then more came up for Friday, so I extended again; and then Monday, too. Now people are even telling me, “Hey, we’re in energy too—come down to Houston and let’s chat.” So our schedule for the entire week after the conference is completely packed with meetings. It’s pretty interesting with AI engineers—a lot of them are feeling overwhelmed, and that pressure comes straight from management. Why? Because management keeps saying, “We don’t need a team, you are a team of one.” And suddenly it’s, “Hey, now I have to build out this entire system on my own,” whereas before that would easily take ten people. All right, we’re here in the studio. This is my final podcast episode at AIE, joined by Anekha from Moshi. Welcome! Thank you so much.
1:01 Did I pronounce your name right, Anekha? Yes, spot on. Your background caught my eye because I used to be a trader myself, and now you’re a founder. Yes. You mentioned a bit of your background to me earlier, and I’d love to hear it firsthand. As far as I know, this is your first time attending AIE. How did you find out about us? How was the experience giving your talk? And what happened afterwards? Well, first off, huge thanks to everyone at AIE, because the whole experience has been incredible. And thank you so much for having me on the podcast. I’m thrilled to share my story. Also, I think we’re living through an era moving at breakneck speed, where practically everyone I talk to feels overwhelmed. So being able to reflect on my journey is really grounding—my undergrad was in statistics and mathematics, which is essentially what everyone calls “AI” today. Getting to discuss that evolution, and especially connecting with a fellow former trader, is super exciting.
2:02 Trading war stories, exactly. Did you do financial math, like stochastic calculus? Yes, I did. My undergrad was in Mathematics, Operational Research, Statistics and Economics (MORSE) at the University of Warwick. Back then, people would react like, “Oh gosh, that is so geeky and nerdy.” I was almost embarrassed to tell people what I studied. And now, everyone’s like, “Oh wow, machine learning! AI!” Exactly. And then derivatives. I covered all of that. So my journey started at Warwick in England, and right around then, I knew I wanted to go into trading. I don’t know what your path into trading was, but I craved a fast-paced environment where I could work with data—I just loved data. When I read about what trading was, I couldn’t see myself doing anything else. So I set my sights on it. I landed my first internship in investment banking and trading, followed by an internship at JPMorgan, where I officially kicked off my career.
3:04 I started out on the emerging markets desk, and later moved over to commodities, trading derivatives. So the bulk of my career was spent in energy derivatives. Coming from a data background, I was always inclined to break the market down into its component parts. In commodities, for example, so much hinges on fundamentals, news flow, and what’s happening in the physical market. Because of that, I was always fascinated by building systems from the ground up. As a trader, I saw that dynamic play out: first the data drops, and then real-world events unfold right after. That was when I left JPMorgan, with the intention of starting a company and becoming a founder. But I wanted to build a data company. So I applied for and received a Fulbright scholarship to come to the US, because your venture ecosystem over here is just incredible. And just casually walked away with a Fulbright scholarship.
4:06 Yeah, there’s actually a funny story behind it—I was taking a yoga class in London at the time. Here’s how it went: my parents are in the jewelry business, and before starting my company in college, I had also worked on a computer vision startup on the side…