Catalyst raises $30M seed from Sequoia for AI trading agents aimed at retail investors

Catalyst, a startup building AI trading agents for retail investors, raised a $30 million seed round led by Sequoia Capital, Fortune reported on 8 October 2026. Users describe a goal in plain language, and the agents build and execute a trading strategy, with a human approving each action. Fortune and Catalyst's own announcement do not disclose a valuation.

How much did Catalyst raise and who invested?

Catalyst raised $30 million in a seed round led by Sequoia Capital. Fortune's Term Sheet newsletter broke the news on 8 October 2026, writing that "Sequoia recently led Catalyst's $30 million seed round." Catalyst's own announcement, as embedded in Pulse 2.0's coverage, also calls it a $30 million seed round.

The two accounts name different co-investors. The table separates what each source says.

Jump, Lux Capital, AntiFund and Peak XV appear in both. Coinbase and Premji Invest appear only in Fortune's list, according to Fortune. Selini Capital and SF1 appear only in the company's post. Dealroom and Crypto Briefing repeat Fortune's list.

Sequoia partner George Robson is the only investor quoted by name. Per Pulse 2.0, he said it was "our privilege to be their early partner and lead Catalyst's seed round."

What does Catalyst do?

Catalyst builds AI agents that turn a retail investor's stated goal into a trading strategy and then execute it. Dealroom, summarizing Fortune, says the agents turn natural language into a trading strategy, "with a human required to approve each move."

Catalyst's announcement text, as reproduced by Pulse 2.0, describes the product this way: users describe a thesis, and the agents gather and analyze market signals, find execution paths across platforms and venues, and help users backtest and deploy strategies. The company says the product includes limits and safety controls.

Cofounder Dylan Iskandar told Fortune the long-term idea is that people "simply state an intent" and the agents handle everything end to end. Sequoia's Robson, per Fortune, described the agents as "a suit of armor as you navigate the financial world."

What traction has Catalyst shown?

Catalyst says a pilot generated hundreds of millions of dollars in trading volume over a couple of weeks. This is a company claim, and Fortune attributes it to the company. The pilot focused on power traders, a group likely to trade far more than a typical retail user.

Fortune also reports that users are now being let off the product's waitlist, with wide availability expected later. No date was given.

Volume is also not revenue or users. The sources do not say how many people took part in the pilot, which brokerage executed the trades, or how Catalyst makes money.

Who founded Catalyst?

Justin Zheng and Dylan Iskandar founded Catalyst in 2025. Fortune gives their ages as 25 and 21. The table summarizes their backgrounds as reported.

Fortune reports the two met as teenagers in 2019 at a conference run by economist Tyler Cowen. According to Pulse 2.0, the wider team has worked at Palantir, Scale AI, Citadel, Citadel Securities, Hudson River Trading and Jump Trading. That list comes from the company's announcement.

How much has Catalyst raised in total?

The only funding round disclosed in the coverage is this $30 million seed. No earlier round is reported.

Valuation is not disclosed.

What rules apply to AI agents that trade?

The sources reviewed do not say which licenses or registrations Catalyst holds, or which firm routes its orders. That is the first question regulators and users would ask. What can be said is how the main broker-dealer self-regulator treats AI agents.

FINRA's 2026 Annual Regulatory Oversight Report, released on 9 December 2025, added a new section on AI agents. It defines them as "systems or programs that are capable of autonomously performing and completing tasks on behalf of a user." FINRA lists risks including:

• "AI agents acting autonomously without human validation and approval."
• "Agents may act beyond the user's actual or intended scope and authority."
• Multi-step reasoning that can make outcomes "difficult to trace or explain, complicating auditability."

FINRA suggests that member firms consider where to put human-in-the-loop oversight, how to track agent actions and decisions, and how to set guardrails that limit agent behavior. The wording is permissive ("may wish to consider"), not a rule. The report is addressed to FINRA member firms, and the sources do not say whether Catalyst is a member or partners with one.

Catalyst's human-approval design lines up with the first of those concerns on paper. Whether it holds in practice depends on how the approval step works.

Why is the gambling question being asked?

Fortune's reporter raised it directly: with prediction markets and crypto now mainstream, a product that makes trading easy invites worries about misuse. Cofounder Justin Zheng answered in Fortune, "We're not here to build a casino" and "We're not here to build a gambling app."

Zheng also told Fortune the company is building user education tools that explain safety and risks, personalized to each user. Iskandar said, "We want to build safe, trustworthy finance." These are statements of intent. The sources do not describe the education tools in detail or show how they perform.

Crypto Briefing raised a related point in its own commentary: it is unclear whether retail users will meaningfully review each AI recommendation or simply tap confirm. That is the outlet's view, not a finding.

What's still unclear?

Several basic facts are missing from the public record reviewed.

• Investor list. Fortune and Catalyst's own post name different co-investors. Coinbase and Premji Invest appear only in Fortune. Crypto Briefing says neither Coinbase nor Catalyst has announced any integration, according to that outlet.
• Valuation and use of funds. Neither is stated in Fortune or Pulse 2.0.
• Execution and licensing. The sources do not identify the brokerage, custodian or regulatory status behind the trades.
• Business model. Fortune does not report fees or how Catalyst earns revenue.
• Pilot volume. The "hundreds of millions" figure is the company's claim, with no user count or breakdown.
• Weakest point of the thesis. A human-approval step reduces the risk of an agent acting on its own, but a retail user clicking approve on a plan they do not fully understand is a different risk. The sources do not show how Catalyst addresses it beyond education tools.