CEO Fredrik Haga says the firm is well capitalized and refocusing on Dune MCP, an AI agent that builds dashboards without SQL, amid a wave of AI-led layoffs.
Alpha Score of 51 reflects moderate overall profile with moderate momentum, strong value, moderate quality, poor sentiment.
Crypto data firm Dune has cut 25% of its workforce, with CEO Fredrik Haga announcing a restructuring designed to sharpen the company’s focus on AI-powered data tools and the rising wave of institutional clients moving onchain. The reduction places Dune alongside similar AI-driven headcount moves at Block and Crypto.com, creating a readthrough for the entire crypto data infrastructure sector that extends well beyond one company’s internal reorganisation.
Haga’s statement this week was unambiguous: the layoffs are a pivot toward two specific growth bets, not a retreat from financial pressure. “We remain well capitalized, excited about the future, and committed to our mission of making crypto data accessible,” he wrote. The immediate signal is that Dune will commit engineering and capital to a product, Dune MCP, that lets AI agents build dashboards without SQL knowledge. The deeper implication is that the competitive moat in crypto data is shifting from raw data aggregation to integrated, agent-ready infrastructure.
Haga’s announcement was direct about the rationale. He told the company:
The departing employees were described as exceptional, with Haga going out of his way to recommend them to other hiring teams. No timeline was given for rebuilding headcount. The smaller workforce is expected to move faster on the two priorities Haga identified: AI integration and the institutional onchain opportunity. The company, founded in 2018, has already built a full-stack data infrastructure covering ingestion, quality assurance, storage, cleaning, normalisation, and querying. The restructuring is meant to push that stack deeper into automated, agent-accessible territory rather than merely sustaining the status quo.
Haga’s explicit statement that Dune remains well capitalised reframes the layoff as an offensive bet. A company cutting from weakness would preserve cash and shelve ambitious product launches. Dune, by contrast, is putting its remaining resources behind an AI-first institutional strategy. The logic is that a leaner, more specialised team can ship faster when the product roadmap is narrowed to a few high-conviction bets. The risk is that cutting 25% of staff may also remove institutional knowledge, community relationships, and the capacity to maintain the thousands of existing dashboards that analysts depend on. Dune’s brand strength was built on a broad user base of crypto-native analysts. The pivot toward AI and institutions could strain that relationship if the legacy product stagnates while the new tools mature.
The centrepiece of the AI strategy is Dune MCP, a product Haga says occupies a position no competitor currently matches. The tool allows teams and AI agents to build dashboards and execute data workflows without any knowledge of SQL or data infrastructure. Haga elaborated:
The end-to-end stack Haga references includes several layers that historically required separate tools or internal engineering:
Dune MCP abstracts all of that behind an interface an AI agent can call directly. The commercial logic is straightforward: the addressable market expands from SQL-literate analysts to anyone who can describe a data question in natural language or delegate it to an AI assistant. That dramatically widens the funnel, particularly among institutional clients who may have deep capital allocations but no in-house blockchain data engineers.
Most crypto data providers offer analytics layers that sit on top of raw data others have indexed. Dune’s claim is that owning the entire pipeline–from ingestion to query resolution–allows for tighter integration with AI agents, faster iteration on data quality, and a lower incremental cost per new blockchain supported. The company has operated through multiple market cycles over eight years, outlasting several competing data platforms that shut down or scaled back. That durability gives weight to the assertion that the infrastructure moat is real.
For competitors, the challenge is that replicating the full stack requires significant engineering investment. The window to do so may narrow further as institutional money starts flowing onchain. A firm that only provides a SQL frontend or a dashboard library cannot easily compete with a system an AI agent can call programmatically. The read-across is that pure-play analytics firms face margin pressure unless they too build agent-ready interfaces. The shift coincides with a broader architectural move toward modular blockchains, where execution, settlement, and data availability layers are separated. That fragmentation multiplies the number of onchain data sources and makes a unified data provider more valuable. AlphaScala’s analysis of developers moving away from monolithic chains details how this trend creates a fractured data environment–exactly the environment where Dune’s end-to-end stack, updated continuously, can act as the aggregation point that simplifies complexity for institutional users.
Haga named institutions coming onchain as the second pillar of Dune’s strategy, alongside AI. The shift is material: traditional assets–currencies, stocks, bonds, commodities–are increasingly being tokenised or represented on blockchain rails. Financial firms that want exposure, or that need to monitor onchain activity for compliance and risk management, require data infrastructure as reliable as the Bloomberg Terminal but built for decentralised networks.
Dune is investing in its data layer and in dedicated client services to serve that market. The institutional use case differs from the retail crypto analyst in specific ways:
A full-stack provider that already ingests and normalises data from dozens of chains is in a better position to meet those demands than a startup that has to build the pipeline from scratch. The AI layer then becomes the differentiator: instead of hiring a team of analysts to write and maintain SQL scripts, an institution can plug an AI agent into Dune MCP and generate dashboards for regulatory reporting, portfolio monitoring, or market surveillance on demand.
Key insight: Dune’s pivot is not merely about replacing dashboards with chatbots. It is about becoming the data infrastructure layer that AI agents and institutional workflows both consume directly. That model, if executed well, creates a switching cost that is hard for fragmented competitors to replicate.
The layoffs at Dune, Block, and Crypto.com form a pattern that spans payments, exchange, and data infrastructure subsectors. The common thread is the deployment of AI tools that reduce the need for human labour in roles ranging from customer support to software engineering to data analysis. Each firm insisted the cuts were proactive, not reactive.
Block’s February 2026 reduction, which eliminated close to 4,000 positions, was the largest in absolute terms and was explicitly framed as an efficiency gain from AI. Crypto.com’s March reduction was smaller in headcount but symbolically potent because the exchange had grown rapidly during the previous cycle. Dune’s layoff is the smallest of the three in absolute numbers, yet it carries sector-specific weight because Dune sits at the infrastructure layer that many other crypto businesses depend on.
Critics in each case have questioned whether AI is the primary driver or a convenient framing for broader cost-reduction efforts. Haga tied Dune’s decision directly to product strategy, however, not to a capital shortfall. The firm’s insistence that it remains well capitalised distinguishes the move from a downsizing born of necessity. The crypto sector is undergoing an AI-driven efficiency push that mirrors what has happened in broader technology. The firms that move first to restructure around AI agents may gain a cost advantage. They also risk losing the domain expertise that only experienced human analysts possess. The balance between automation and human judgement will determine whether these cuts prove prescient or premature.
The layoff and AI pivot at Dune send a signal to every other crypto data provider that the competitive landscape is changing under their feet. The readthrough is not limited to direct rivals; it extends to any firm whose value proposition depends on human-led analytics or siloed data sets.
Readthrough 1: Pure analytics firms face margin pressure. Companies that sell dashboard subscriptions or custom research reports may find their services undercut by AI-generated insights that are cheaper and faster to produce. If Dune MCP can answer the same questions a research shop addresses, the value of human curation diminishes.
Readthrough 2: The moat moves to data engineering, not data access. Dune’s argument is that owning the ingestion and normalisation layer is the hard part, and AI agents only work well when the underlying data is clean and consistently formatted. Competitors that rely on third-party data sources may struggle to deliver the same query reliability.
Readthrough 3: Institutional procurement cycles favour integrated stacks. Financial firms buying data infrastructure prefer vendors that can cover multiple use cases under a single contract. A provider that offers the data pipeline plus an AI agent layer reduces procurement complexity and may capture larger, stickier contracts.
Readthrough 4: Workforce strategy becomes a signal of product confidence. Dune’s decision to cut staff while well capitalised signals that management believes the AI tools are ready for production. If the products fail to deliver, the market will treat the layoff as a misstep rather than a strategic acceleration. Competitors watching from the sidelines can use this as a live case study: if Dune gains share, the pressure to follow increases.
The sector-wide impact on crypto market activity is indirect, yet real. Better data infrastructure tends to increase market efficiency, transparency, and institutional participation. As documented in ongoing crypto market analysis, the entry of institutional-grade data tools is a necessary condition for broader asset class adoption.
Haga closed his statement with a three-word line: “The Data Must Flow.” The phrase captures the operational priority–that the dashboards and queries millions of users rely on must remain available and reliable through the restructuring. It also hints at the execution risk embedded in the pivot. AI agents that build dashboards automatically need to produce results that are not only fast but accurate. An error rate that a human analyst would catch before publishing could, in an agent-driven workflow, propagate directly into investment decisions.
Practical rule: AI-driven headcount reductions are a bet that the remaining team can multiply output. If the product does not ship faster and with measurable quality, the market will price in execution failure.
The Dune MCP product must demonstrate real adoption metrics–number of agents connected, queries served, institutional contracts signed–within two to three quarters. Without that, the market will question whether the 25% headcount reduction sacrificed maintenance capacity for a product that is not yet mature. A decline in existing dashboard usage or community engagement would further erode confidence.
The flip side is that if Dune ships the MCP product quickly and signs recognisable institutional clients, the company could emerge from this transition with a structurally lower cost base and a revenue profile tilted toward higher-value, recurring contracts. The data infrastructure play in crypto has historically been a low-margin, high-volume grind. An AI layer that captures enterprise budgets could change that equation. The coming quarters will test whether Dune’s commitment to an AI-first institutional data stack is a durable competitive shift or a high-stakes gamble that leaves the company stretched between community users and Wall Street demands.
Prepared with AlphaScala research tooling and grounded in primary market data: live prices, fundamentals, SEC filings, hedge-fund holdings, and insider activity. Each story is checked against AlphaScala publishing rules before release. Educational coverage, not personalized advice.