--- url: 'https://qubit.capital/blog/us-seed-weekly-funding-roundup-week-5-june-2026' title: 'US Seed Weekly Funding Roundup (Jun 22-29, 2026): $65.0M Raised Across 2 Deals' author: name: Sahil Agrawal url: 'https://qubit.capital/blog/author/sahil' date: '2026-06-29T06:18:27+05:30' modified: '2026-06-29T11:25:41+05:30' type: post summary: 'US seed funding hit $65M this week across 2 AI deals: Ornn''s $33M GPU compute marketplace and Hang Ten''s $32M AI-native IT services play.' categories: - Weekly Funding Roundup image: 'https://qubit.capital/wp-content/uploads/2026/06/featured-us-seed-67954.webp' published: true --- # US Seed Weekly Funding Roundup (Jun 22-29, 2026): $65.0M Raised Across 2 Deals US seed funding stayed concentrated and AI-heavy this week, with two deals splitting $65.0 million almost down the middle. [Ornn](https://ornn.com) took $33 million to turn GPU compute into a tradable commodity, and [Hang Ten Systems](https://hangten.ai) raised $32 million to rebuild enterprise IT services around AI agents. Both rounds went to founders who already know their markets cold, and both target infrastructure problems created by the AI boom itself. The seed numbers look small next to what later-stage US startups pulled in. Series A companies raised $657.0 million across six deals the same week, and Series B and beyond added $226.0 million across three more. That gap is normal, but the seed deals here are not typical first checks. Ornn’s round is a large follow-on with a16z leading, and Hang Ten launched with a $32 million seed and a founder who used to run Infosys. Investors are writing seed-stage tickets at growth-stage conviction when the team and thesis are clear. Weekly Funding Roundup JUN 22-29, 2026 $65M TOTAL RAISED 2DEALS CLOSED 100%SEED $32.5MAVG DEAL SIZE USTOP REGION BY STAGE Seed$65M100% BY SECTOR OrnnAI infrastructure and capital markets$33M Hang Ten SystemsEnterprise AI services$32M Table of Contents - [1. Ornn Raises $33M For GPU Compute Trading](#1-ornn-raises-$33m-for-gpu-compute-trading) [Deal Overview](#deal-overview) - [Investor Profile](#investor-profile) - [Company and Leadership](#company-and-leadership) - [Problem and Opportunity](#problem-and-opportunity) - [Product and Technology](#product-and-technology) - [Use of Proceeds and Vision](#use-of-proceeds-and-vision) - [Market Context](#market-context) - [2. Hang Ten Systems Raises $32M For AI IT Services](#2-hang-ten-systems-raises-$32m-for-ai-it-services) [Deal Overview](#deal-overview-1) - [Investor Profile](#investor-profile-1) - [Company and Leadership](#company-and-leadership-1) - [Problem and Opportunity](#problem-and-opportunity-1) - [Product and Technology](#product-and-technology-1) - [Use of Proceeds and Vision](#use-of-proceeds-and-vision-1) - [Market Context](#market-context-1) - [Lessons For Founders](#lessons-for-founders) ## 1. Ornn Raises $33M For GPU Compute Trading ### Deal Overview - Stage: Seed (large follow-on) - Sector: AI infrastructure and capital markets - Geography: United States - Round size: $33 million - Lead investor: Andreessen Horowitz, with a16z crypto, Galaxy Ventures, and SV Angel ### Investor Profile a16z led the round alongside its crypto arm, which signals the firm sees Ornn as both an AI infrastructure bet and a financial markets bet. Galaxy Ventures brings commodity and digital-asset trading depth, and SV Angel adds early-stage credibility. The backing matters because Ornn is building a market, and markets need trusted sponsors. Having a16z on the cap table helps Ornn recruit the cloud providers and institutional buyers it needs on both sides of the exchange. ### Company and Leadership Ornn was founded in 2025 by two MIT graduates, Kush Bavaria (CEO) and Wayne Nelms (CTO). Nelms previously traded equity options at Susquehanna International Group, so the team mixes quantitative trading with systems engineering. The company raised a $5.7 million seed in October 2025 to build what it called the world’s first compute futures exchange. This $33 million round, announced June 24, 2026, is a fast and much larger expansion of that first bet. ### Problem and Opportunity Companies buy GPU compute through private, opaque deals. Prices vary widely, contracts are hard to compare, and there’s no clean way to hedge cost. CFOs and lenders can’t underwrite spending they can’t price. Ornn wants to fix that by treating compute like oil or electricity, a commodity with public benchmarks, secondary markets, and hedging tools. ### Product and Technology Ornn Compute is a marketplace that aggregates GPU capacity from clouds and neoclouds, with transparent pricing, secondary transfers, and on-demand sublets. Its flagship benchmark, the Ornn Compute Price Index (OCPI), is built from live negotiated trades across H100, H200, B200, B300, and RTX 5090 hardware. OCPI is already distributed on the Bloomberg Terminal. On top sit cash-settled futures, swaps, and residual-value protection. In May 2026, Intercontinental Exchange announced a partnership to launch GPU compute futures referenced to OCPI. The moat is the data. OCPI prices come from cleared trades, not scraped list prices. The marketplace feeds the index, and the index’s credibility drives volume back to the venue. ### Use of Proceeds and Vision The $33 million funds hiring, marketplace liquidity, broader index coverage, and the capital-markets product suite, including the ICE futures launch pending CFTC approval. The thesis is the financialization of compute. Ornn’s tagline is “financial markets for the world’s most important resource,” and leadership frames compute as an input that investors and lenders need to underwrite. ### Market Context The AI GPU rental market is projected to grow from about $3.3 billion in 2023 to roughly $34 billion by 2032. CoreWeave alone is guiding to $12 billion to $13 billion in 2026 revenue. Ornn’s closest competitor is Silicon Data, whose H100 Rental Index backs a rival CME Group futures effort. That makes this a two-horse race: ICE plus Ornn against CME plus Silicon Data. On the marketplace side, Ornn aggregates across CoreWeave, Lambda, Crusoe, and Nebius rather than owning hardware. ## 2. Hang Ten Systems Raises $32M For AI IT Services ### Deal Overview - Stage: Seed - Sector: Enterprise AI services - Geography: United States (Palo Alto, California) - Round size: $32 million - Lead investor: Mayfield, with Aramco Ventures ### Investor Profile Mayfield led the round, and Aramco Ventures joined as a strategic backer. Aramco’s investors pointed to Hang Ten’s mix of frontier AI experience and decades of enterprise delivery as the reason they wrote the check. Yahoo co-founder Jerry Yang sits on the board, which adds Silicon Valley network and credibility to an already senior team. ### Company and Leadership Hang Ten was founded in 2026 and is led by Dr. Vishal Sikka, the former CEO of Infosys and a former SAP Executive Board member. He’s targeting the same IT services model he once ran at scale. The founding team reunites veterans from Sikka’s prior ventures: Navin Budhiraja (CTO), Sanjay Rajagopalan (Chief Design Officer), and Tao Liu (SVP of Forward Deployed Engineering). The company is separate from Sikka’s earlier firm, Vianai Systems. ### Problem and Opportunity Sikka’s view is blunt. Every enterprise will be reshaped by AI, but most are stuck at the starting line, and the gap is widening. Traditional IT services scale by adding people, which keeps costs high and timelines long. Hang Ten wants to close that adoption gap with an approach that doesn’t grow linearly with headcount. ### Product and Technology Hang Ten sells an AI-native operating model that lets large enterprises build, change, and run their software continuously. It rests on three pillars: agentic code generation that builds and operates software, a reusable skills library that compounds across projects, and an FDE Bench of Forward Deployed Engineers who embed domain expertise into delivery. It targets enterprise transformations, finance, HR, and new product development, and it sells software as an outcome rather than a license. The economics are the point. As the skills library grows, leverage rises and marginal cost falls, unlike headcount-driven consulting. Early customers include Siemens Gamesa and Fresenius, helped by Sikka’s relationships across SAP, Oracle, and Infosys. ### Use of Proceeds and Vision The $32 million funds hiring across delivery, engineering, sales, and leadership, plus expanding enterprise engagements and building out the FDE Bench. The ambition is to challenge the roughly $250 billion traditional IT services and outsourcing industry from an AI-native angle, taking aim at the labor-arbitrage model that built Infosys and its peers. ### Market Context Hang Ten sits between enterprise AI adoption and the agentic software wave, going after incumbents like Infosys, TCS, Accenture, Wipro, and Cognizant. On the tooling side, agentic coding includes Cognition’s Devin, Cursor, GitHub Copilot, and Replit, but Hang Ten sits a layer up by selling outcomes and embedded engineers rather than an IDE. Incumbents are reacting through partnerships, such as Infosys working with Anthropic and OpenAI. That response shows how fast AI code generation is cutting the cost and time of building enterprise software. ## Lessons For Founders - Operator credibility moves seed rounds. A former Infosys CEO and an ex-Susquehanna trader raised $32M and $33M respectively because investors trusted the team to execute on hard markets. - Distribution can be a moat. Ornn put its price index on the Bloomberg Terminal, giving it instant trust with the CFOs and lenders it needs as customers. - Attack economics, not just features. Hang Ten’s pitch isn’t better code, it’s a delivery model where cost falls as the skills library grows, breaking the headcount math of legacy consulting. - AI is creating its own infrastructure gaps. Both winners build on top of the AI boom, one pricing the compute that powers it, the other rebuilding the services layer around it. - A seed round can carry growth-stage conviction. When the thesis and team are clear, investors will write large early checks rather than wait for proof.