Jump Trading has hired the founding team of Itô, a startup that trades access to computing power, Bloomberg reported on 13 October 2026, citing a statement on social media. In that statement, published on X on 12 October 2026, Itô co-founder Affaan Mustafa wrote that “the Itô team has joined Jump Trading to lead its compute trading business.”
Mustafa wrote that he and co-founder Alejandro will lead a new internal compute trading desk at the firm, and that “We’re active and trading in it today”.
Business Pill · TAKE-OR-PAY
A short explainer of take-or-pay: committing to pay for capacity whether or not you use it. It teaches the general idea only and says nothing about any company in this story.
The key insight: As we read it, the desk is being built around a timing gap the founders describe: capacity contracts that commonly start at 3 to 5 year commitments with around 30% paid upfront, sold to companies whose revenue, in Mustafa’s words, is often month-to-month. No terms of the desk itself have been disclosed.
Update, 16 October: DRW joins, and Jump says it will trade on its own balance sheet. Bloomberg reported on 16 October that Jump Trading and DRW are committing traders and capital to the emerging market for contracts tied to computing power. Jump’s dedicated team, which includes Affaan Mustafa and Alejandro Soumah Fernandez from Itô Markets, will use Jump’s own balance sheet to take positions and plans to trade over-the-counter deals, perpetual futures and prediction bets. Jump expects to make markets in the compute futures planned by CME and Silicon Data, which remain subject to regulatory approval, as well as rival contracts from Intercontinental Exchange and Ornn.
DRW plans to trade compute through three teams, two existing and one new, and will trade the Silicon Data products, Bloomberg said. DRW founder Don Wilson co-founded Compute Exchange and invested in Silicon Data; Jump is an investor in both Silicon Data and Ornn. “Right now, the market is very opaque, and there isn’t good price transparency,” Wilson told Bloomberg.
Jump Trading president and CIO Dave Olsen said hedge funds, dealers and other traders could dominate futures trading for “a prolonged period” before AI labs and neoclouds use the contracts, because those buyers will “stay on the sidelines until they see the cash settled products track their procurement experience.” As we read it, the desk now has a named counterpart at DRW and a stated plan to warehouse risk, while Bloomberg’s article gives no trading volumes or terms for the desk itself.
Update, 20 October: Commodity brokers prepare for US compute futures. Some top commodity brokerage firms, including ADM Investor Services, BGC Group and Nanhua USA, are preparing for possible approval by the Commodity Futures Trading Commission of compute futures trading in the US, Bloomberg reported on 20 October. The firms sit between the exchanges and customers, executing orders for hedge funds, commodities traders, commercial companies and other institutional investors, Bloomberg wrote.
For context, CME Group and Silicon Data announced on 11 August plans to launch two compute futures contracts, Silicon Data H100 Rental Index Futures and Silicon Data B200 Rental Index Futures, listed under NYMEX rules, on 5 October 2026 “pending regulatory review.” As we read it, the brokers are the plumbing that would let hedge funds and commercial users reach those contracts once trading is cleared.
Mustafa’s post describes how Itô began: the founders first set out to use “new primitives and contract markets to hedge corporate risk that the current system simply could not support.” He wrote that they spoke to “over a thousand funds, corporations, VCs, and people in every industry”, from reinsurers and AI CFOs to startup founders and hedge funds.
The post says the founders had also been end buyers of compute themselves, “training and serving models across providers.” It says Alejandro came from “creating credit baskets at GS”.
Asked why they would run the business as a desk inside a larger firm, Mustafa listed “Good credit, risk controls, deep trading experience, hardware expertise, talent, resources, and freedom from the constraints of equity rounds or private-credit debt”, and wrote that “at Jump they do” come together. The post gives an inbound address for the desk, compute@jumptrading.com.
The contract terms Itô co-founder Affaan Mustafa describes in his 12 October 2026 post: capacity contracts that commonly start at 3 to 5 year commitments with around 30% paid upfront, and, by his account, variable costs that often made up 60+% of company expenditures for the layer above the models.
The Contract Terms the Post Describes
The post sets out the problem the desk is built around. Mustafa wrote that “Capacity contracts commonly start at 3 to 5 year commitments with around 30% paid upfront.”
Against that, he wrote, the buyers’ revenue “is new, usage-based, and often month-to-month, and they need the compute before the first dollar comes in.” He added that secondary-market prices for the most in-demand chips are spiking and volatile.
The post locates the gap in the middle of the market: a “fast-growing cohort of late-seed to Series C companies using venture capital and equity to fund compute”, whose clusters “can carry TCVs in the low to mid hundreds of millions for the mid-market at times.” TCV is total contract value.
It also says that variable costs “often made up 60+% of company expenditures for the layer that sits above the models.”
What Itô’s founders state in their 12 October 2026 posts about the new desk, and what the texts we read do not disclose: a statement from Jump Trading itself, deal terms, desk size and trading volumes.
Where This Sits
Bloomberg framed the hire as market makers moving “into an emerging asset class underpinning the AI boom.” Mustafa’s GitHub profile now lists him as “Compute Trading Lead @ Jump Trading” and describes Itô as “(acq.)”.
Mustafa is also due on the main stage at the AI Engineer conference on 14 October 2026 with a talk he titled “The Dark Pools of Compute”, which he describes as a look at how compute is bought and sold and what sets its price.
The Structural Read
Mustafa’s post puts the gap in the middle of the market rather than at the top: late-seed to Series C companies funding compute with venture capital and equity, with clusters that can carry total contract values in the low to mid hundreds of millions.
As we read it, the founders’ own description of that cohort is the clearest statement of the problem: “Too small to take on significant debt, too big for meaningful subsidies, and facing more demand than they could afford to serve.”
The post says the founders chose to run the business inside a larger firm for its credit, risk controls, trading experience and hardware expertise, and for freedom from the constraints of equity rounds or private-credit debt.
Affaan Mustafa, Itô co-founder, 12 October 2026
“Capacity contracts commonly start at 3 to 5 year commitments with around 30% paid upfront.”
Three Implications
JUMP TRADING Mustafa wrote that the Itô team will lead a new internal compute trading desk at Jump, and that it is already trading; Jump itself has not published a statement that we found.
MID-MARKET AI COMPANIES The post describes late-seed to Series C companies whose clusters can carry total contract values in the low to mid hundreds of millions, paying upfront for compute before revenue arrives.
THE COMPUTE MARKET Bloomberg framed the hire as market makers moving into an emerging asset class underpinning the AI boom; Mustafa wrote that the market is forming and that the desk wants to help shape it.
The framework puts it this way: “For two years, AI was a CIO line item and a CEO talking point. In 2026 it became a CFO problem, because the economics finally turned legible — and unforgiving.”
As we read it, a trading firm building a desk for multi-year compute contracts is that CFO problem turning into a market: the cost of compute becomes something to price, hedge and trade rather than only to pay.
What Is Not Established
We did not find a statement from Jump Trading itself, and no deal terms, desk size or trading volumes have been disclosed in the texts we read. Bloomberg’s article body was not readable to us; we read only the summary it posted on X, and the facts above come from the founders’ own posts.
Whether Itô was acquired or its team hired is described only by Mustafa’s profile wording “(acq.)”; Bloomberg’s summary says Jump hired the founding team.
Business Engineer Framework
The CFO’s Guide to the Token Economy
A Business Engineer guide to how AI compute and token costs land on the finance side of the business.
Bloomberg reported on 13 October 2026 that Jump Trading has hired Itô’s founding team, citing a statement on social media. In that statement, co-founder Affaan Mustafa wrote that the team will lead a new internal compute trading desk at Jump, which he said is already trading, in a market where, by his account, capacity contracts commonly start at 3 to 5 year commitments with around 30% paid upfront.
A note on sourcing. Bloomberg reported this story on 13 October 2026; its article body was not readable to us, so we read only the summary it posted on X. The 16 October update comes from Bloomberg’s 16 October article (Muyao Shen), read in full. The facts in this piece come from the founders’ own posts: Itô co-founder Affaan Mustafa’s X article “We’ve joined Jump Trading” of 12 October 2026, read in full, his post announcing his AI Engineer talk, and his GitHub profile, all read on 13 October 2026.
We did not find a statement from Jump Trading itself. Nothing here is a forecast, and nothing here is financial or investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
