MEGA, the company behind the Ethereum scaling network MegaETH, cut 20 percent of its staff on November 14 as it pivots toward an AI compute marketplace launching in December, after decentralized finance activity on the network collapsed from a peak above $240 million to just $8.8 million and its token fell more than 70 percent this year.
Co-founder Shuyao Kong announced the cuts on X on Friday, saying they hit staff working on the blockchain itself, the wallet, and the broader ecosystem team. The founders and core engineering staff are staying put. It is a familiar shape for crypto this year: raise a huge round, build fast, watch usage dry up, then point everything at AI and hope the market rewards the pivot.
A fast reversal for a network that raised $450 million
MegaETH did not start as a struggling project. MegaLabs, the team behind it, raised $20 million in 2024 in a round led by Dragonfly and backed by Ethereum co-founder Vitalik Buterin. Then in late 2025, MegaETH pulled in $450 million in an oversubscribed token sale, one of the larger raises in the layer-2 space that year.
The network was built to make blockchain apps like trading platforms and games run faster than they could on Ethereum’s base layer. That pitch worked well enough to draw hundreds of millions of dollars in deposits at its peak. It has not worked well enough to keep them there.
Where the cuts landed
Kong did not give an exact headcount for the layoffs, only confirming that the blockchain, wallet, and ecosystem teams absorbed the cuts while engineering stayed intact. Earlier in the week, she had already signaled the shift was coming, writing that “staying a chain-only company felt wrong” given what she described as the bigger opportunity in front of the team.
That opportunity is the restructuring behind the move. The MegaETH Foundation has rebranded as MEGA, and the new entity has committed $30 million from its treasury to build products on top of the network, with AI services first in line.
The numbers behind the decision
The pressure on MegaETH to find a new growth story is easy to see in the data. Total value locked in its DeFi ecosystem, tracked by DeFiLlama, has dropped from more than $240 million at its height to $8.8 million now. The MEGA token itself has shed more than 70 percent of its value this year, according to CoinDesk price data.
For a network that raised nearly half a billion dollars on the promise of fast, cheap blockchain execution, that kind of drop is hard to spin as a temporary dip. It is the backdrop the AI pivot is happening against, not a side note to it.
What the AI marketplace is supposed to do
MEGA’s plan, laid out in a white paper, is to launch a marketplace where customers buy access to AI models from competing providers, all running through MegaETH. The pitch includes letting buyers verify which model actually serves their request and reserve computing capacity at agreed prices ahead of time, a feature aimed at the compute-scarcity problems that have dogged AI infrastructure all year.
The company has not confirmed which of those features will be live when the marketplace launches in December, or how it plans to compete with centralized AI providers that need no blockchain layer at all. Kong has said the new products will run on MegaETH itself, and that the network still plans to finish its decentralization roadmap by the end of March 2027.
The pitch leans on a specific complaint that has followed AI buyers all year: when you pay for access to a model through most APIs, you cannot independently confirm which model actually answered your request, or whether a provider quietly swapped in a cheaper one during a capacity crunch. MEGA’s white paper frames on-chain verification as a fix for that, with usage and billing settled automatically through smart contracts instead of trust in a provider’s own dashboard. Reserved capacity works similarly to how cloud providers sell committed-use discounts, except the commitment and the delivery are both recorded on a public ledger rather than in a private contract.
Why a crypto project is betting on AI right now
MegaETH is not pivoting into AI because the idea came out of nowhere. Verified compute and AI agent traffic have become one of the most active fights in tech this year. As we covered when security researchers found that Meta and OpenAI’s agents now make up close to 90 percent of all AI agent traffic online, money is chasing agent infrastructure hard right now. A blockchain that can verify which model answered a request, and bill for it automatically, is a reasonable bet on where some of that spending could land.
It is also a sign of how thin the margin for error has gotten in crypto infrastructure. Projects that raised huge rounds in 2024 and 2025 are now under pressure to show real usage, not just deposits chasing token incentives. A chain that drew hundreds of millions of dollars in deposits during a bull run, only to see that liquidity drain out almost completely once incentives faded, is not unique to MegaETH. It is close to the default outcome for layer-2 networks that competed mainly on speed and fee discounts rather than a use case nobody else could offer. AI compute, with real recurring demand and a price people are already used to paying elsewhere, is one of the few categories crypto infrastructure teams can point to right now as having genuine, non-speculative buyers.
That pressure shows up elsewhere too: Bitcoin Core disclosed three separate security flaws earlier this month, a reminder that even the most established chains keep finding problems under the hood while newer, less battle-tested networks scramble for a reason to exist at all. MegaETH’s pivot is at least an attempt to find that reason before the money runs out, rather than after.
The bottom line
Cutting a fifth of your staff to chase a product you have not shipped yet is a gamble, not a strategy. MegaETH’s underlying numbers made some kind of change necessary, and an AI marketplace is at least a coherent bet given where compute demand is heading. But a white paper and a December launch date are not proof that customers will show up, and the company is entering a market where centralized AI providers already have the infrastructure, the customers, and a multi-year head start. If the marketplace launches thin and the token keeps sliding, this will look less like a pivot and more like a company trying to outrun its own chart.
The Teck Hustlers Editorial Team covers technology, business, finance and gaming news for readers who want the story behind the headline. Every article is researched, written and edited by our in-house team, checked against company announcements, filings and reputable reporting, and updated when new facts emerge. We focus on what each development means for founders, professionals, investors and everyday users. Have a tip or a correction? Reach us through our contact page.
