MegaETH’s Parent Cuts 20% of Staff as Restructuring, AI Plans Expand

MEGA, the organization behind Ethereum layer-2 network MegaETH, has cut its workforce by 20% after months of internal restructuring, according to statements from MegaETH co-founder Shuyao Kong. The changes reportedly affected employees across multiple functions, including blockchain development, wallet work, ecosystem support, and communications, though Kong did not provide headcount figures or the specific timing of the reductions.

While the layoffs do not appear to have been framed as tied to a single trigger, Kong said the company has reorganized its teams to bring in new capabilities as it shifts toward product development—particularly artificial intelligence services—on the MegaETH network. MEGA has also reportedly committed $30 million toward that initiative, according to its own communications, but Kong did not explicitly connect the staffing changes to the funding.

Key takeaways

• MEGA (MegaETH) reduced staff by 20% following restructuring, with departures spanning blockchain, wallet, support, and communications roles.
• Co-founder Shuyao Kong said the organization’s earliest teams and core engineering group remain committed.
• MEGA is redirecting attention toward AI-focused products built on the MegaETH network.
• MEGA has reportedly committed $30 million to the AI initiative, though the layoffs were not directly linked to that shift.
• MEGA’s plans include launching a public inference market in December, tied to a proposed model for competing AI service providers.

Workforce reduction amid a broader internal reshuffle

In a public post on X, Kong described MEGA’s restructuring as the reason behind the 20% reduction in staff. She did not specify how many employees were affected, nor did she state when the layoffs occurred.

Kong said the reorganization brought “new capabilities” together under a changed team structure. Importantly for those tracking the project’s continuity, she emphasized that founders, early employees, and the core engineering team are still committed to MEGA.

The functions reportedly impacted—blockchain, wallet, ecosystem support, and communications—suggest MEGA is not merely reshuffling one product line but is rebalancing resources across the end-to-end stack: from infrastructure to user-facing tooling and go-to-market.

Strategic shift toward AI products on MegaETH

The workforce changes come as MEGA places greater emphasis on building AI-related offerings on the MegaETH network. Kong said the organization plans to develop products that leverage the underlying chain, including artificial intelligence services.

MEGA has also committed $30 million toward that initiative, though Kong did not directly state that the layoffs were caused by the AI expansion. For investors and users, the key practical implication is that product development priorities are clearly moving toward AI—meaning teams supporting existing infrastructure and ecosystem-facing activities could be reduced while efforts concentrate on new use cases.

That direction aligns with the broader pattern seen across the crypto sector: many networks and startups are attempting to convert protocol-level capabilities into AI-enabling services, either by providing execution, data services, or market mechanisms for compute and inference. However, MEGA’s approach specifically ties AI productization to a market concept described in its whitepaper.

An inference market designed to address provider trust and data concerns

In MEGA’s whitepaper, co-founders Yilong Li and Lei Yang outlined a proposed inference market intended to give users access to competing AI service providers. The premise is that multiple providers would compete to deliver AI inference, offering users options while potentially reducing the risk of relying on a single vendor.

The whitepaper’s framing also targets two recurring concerns in AI outsourcing: whether providers actually use the model they advertise, and how they handle customer data. By creating a marketplace structure around inference access, MEGA aims to make provider behavior more transparent—or at least more inspectable—relative to traditional “single-provider” arrangements.

For builders and users, the most consequential uncertainty is how the market will operationalize those trust and data-handling requirements in practice. The concept suggests that mechanisms may be built into the system’s design, but the available information does not specify the exact enforcement or verification methods.

Public launch target: an inference market in December

Kong said MEGA plans to launch a public inference market in December. That timeline provides a near-term milestone for the project’s AI strategy, moving from whitepaper concepts to a live environment where market participants can test how providers compete and how users interact with the offering.

Earlier coverage from Cointelegraph noted that MegaETH’s mainnet launch was scheduled for Feb. 9 after clocking 35K TPS in testing, underscoring that MEGA has been positioning its network performance as a foundation for higher-throughput use cases. The current shift adds a different layer to that narrative: performance is now being coupled with an application layer intended to support AI inference services and provider selection.

For anyone watching MegaETH, December will likely be an important checkpoint—not only for whether the inference market goes live as planned, but also for whether it can satisfy the trust and privacy concerns it highlights in its whitepaper.

As MEGA restructures and heads into a December launch window, the most important questions for users and investors are straightforward: how the inference market will work in practice, what safeguards it will actually provide around model fidelity and customer data, and whether the staffing changes translate into faster delivery—or gaps—in other parts of the ecosystem.