--- url: 'https://qubit.capital/blog/us-seed-weekly-funding-roundup-week-1-april-2026' title: 'US Seed Weekly Funding Roundup (Mar 27-Apr 3, 2026): $29.5M Raised Across 4 Deals' author: name: Sahil Agrawal url: 'https://qubit.capital/blog/author/sahil' date: '2026-04-06T07:08:19+05:30' modified: '2026-04-06T13:19:02+05:30' type: post summary: 'Four US seed deals totaled $29.5M this week, spanning enterprise AI, quantum interconnects, voice survey agents, and modular robotics hardware.' categories: - Weekly Funding Roundup image: 'https://qubit.capital/wp-content/uploads/2026/04/featured-us-seed-65150.webp' published: true --- # US Seed Weekly Funding Roundup (Mar 27-Apr 3, 2026): $29.5M Raised Across 4 Deals US seed-stage startups raised $29.5M across four deals this week, with enterprise AI and deep tech hardware splitting the action evenly. Two AI-focused companies pulled in a combined $15.2M, while quantum computing and robotics accounted for the remaining $14.3M. Every company this week emerged from stealth or is shipping its first product, a sign that founders are raising on working prototypes rather than slide decks. The seed cohort tells a different story than the bigger checks being written at later stages. US Series A rounds totaled $222.0M across four deals this week, and Series B+ activity was even heavier at $939.0M across seven deals. But seed is where conviction gets tested earliest, and this week’s investors are placing bets on founders with deep operator backgrounds and narrow, defensible wedges. Weekly Funding Roundup MAR 27-APR 3, 2026 $29.5M TOTAL RAISED 4DEALS CLOSED 100%SEED $7.4MAVG DEAL SIZE USTOP REGION BY STAGE Seed$29.5M100% BY SECTOR Whirl AIAI / Enterprise IT$8.9M CavilinQDeep Tech / Quantum Computing$8.8M MiravoiceAI / Market Research$6.3M Anvil RoboticsDeep Tech / Robotics$5.5M Table of Contents - [1. Whirl AI Raises $8.9M to Build AI Agents for Enterprise IT](#1-whirl-ai-raises-$8-9m-to-build-ai-agents-for-enterprise-it) [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. CavilinQ Raises $8.8M to Build Quantum Computing Interconnects](#2-cavilinq-raises-$8-8m-to-build-quantum-computing-interconnects) [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) - [3. Miravoice Raises $6.3M to Replace Call Centers with AI Voice Agents](#3-miravoice-raises-$6-3m-to-replace-call-centers-with-ai-voice-agents) [Deal Overview](#deal-overview-2) - [Investor Profile](#investor-profile-2) - [Company and Leadership](#company-and-leadership-2) - [Problem and Opportunity](#problem-and-opportunity-2) - [Product and Technology](#product-and-technology-2) - [Use of Proceeds and Vision](#use-of-proceeds-and-vision-2) - [Market Context](#market-context-2) - [4. Anvil Robotics Raises $5.5M to Sell Legos for Robots](#4-anvil-robotics-raises-$5-5m-to-sell-legos-for-robots) [Deal Overview](#deal-overview-3) - [Investor Profile](#investor-profile-3) - [Company and Leadership](#company-and-leadership-3) - [Problem and Opportunity](#problem-and-opportunity-3) - [Product and Technology](#product-and-technology-3) - [Use of Proceeds and Vision](#use-of-proceeds-and-vision-3) - [Market Context](#market-context-3) - [Lessons For Founders](#lessons-for-founders) ## 1. Whirl AI Raises $8.9M to Build AI Agents for Enterprise IT ### Deal Overview - **Stage:** Seed - **Sector:** AI / Enterprise IT - **Geography:** San Francisco, United States - **Round Size:** $8.9M - **Lead Investor:** ICONIQ Capital ### Investor Profile ICONIQ Capital leading a seed round is unusual. The firm typically writes checks at growth stage for companies like Datadog, Snowflake, and Adyen. Stepping down to seed signals high conviction in both the founder and the market. For a company emerging from stealth, landing ICONIQ as the lead sets a high bar for follow-on interest. ### Company and Leadership [Whirl AI](https://www.whirlai.com) was founded by Sunny Bedi, who spent 20 years as a CIO and technology executive at Snowflake, NVIDIA, and VMware. He’s not a first-time founder guessing at enterprise pain. He lived it. The company emerged from stealth on March 31, 2026, with this seed as its first announced round. ### Problem and Opportunity Enterprise IT teams sit on years of accumulated system customizations, workarounds, and configuration decisions. Much of that knowledge lives in the heads of people who left the company long ago. When organizations try to modernize their systems, they spend months untangling tribal knowledge before any real work begins. With 71% of organizations modernizing infrastructure to support AI, and companies spending 6-10% of annual revenue on the effort, the bottleneck isn’t technology. It’s understanding what already exists. ### Product and Technology Whirl’s platform continuously ingests metadata from enterprise applications, integrations, and configurations. It converts that raw data into a searchable knowledge base, a real-time map of how systems actually work. Purpose-built AI agents then use that context to help IT teams research, design, and implement changes. The key distinction from competitors like Glean or Moveworks is the depth of metadata ingestion. Whirl doesn’t search documents. It reads the systems themselves. ### Use of Proceeds and Vision Funds will go toward expanding the engineering team, deepening platform capabilities, and beginning early customer deployments. The company positions itself as purpose-built AI agents for enterprise IT, not a generic copilot. The vision is to close the institutional knowledge gap that makes IT transformation slow and expensive. ### Market Context The AIOps market is projected at $18.95B in 2026. Glean sits at a $7.2B valuation, and ServiceNow acquired Moveworks. But none of the existing players focus on IT system modernization through deep metadata ingestion. Whirl is carving a narrow lane with a founder who knows the territory firsthand. ## 2. CavilinQ Raises $8.8M to Build Quantum Computing Interconnects ### Deal Overview - **Stage:** Seed - **Sector:** Deep Tech / Quantum Computing - **Geography:** Cambridge, MA, United States - **Round Size:** $8.8M - **Investors:** QVT Ventures, Safar Partners, MFV Partners, Serendipity Capital, Harper Court Ventures ### Investor Profile Five firms split this round, a common structure for deep tech seeds where no single investor wants full exposure to technical risk. QVT Ventures and Safar Partners both have track records in frontier science bets. The syndicate brings a mix of deep tech and academic-adjacent capital that fits a university spinout well. ### Company and Leadership [CavilinQ](https://www.cavilinq.com) spun out of the University of Chicago in 2025. CEO Shankar Menon completed his PhD there in 2025. The scientific co-founders and advisors, Hannes Bernien at UChicago and Mikhail Lukin at Harvard, are two of the world’s leading researchers in high-fidelity light-matter interfaces. That academic pedigree creates a knowledge barrier that’s hard to replicate. ### Problem and Opportunity Quantum computing is hitting a scaling wall. Individual processors can only hold so many qubits before error rates climb. The industry consensus is that modular, distributed quantum computing, connecting multiple processors into one system, is the path forward. But there’s no standard interconnect layer to make that happen. CavilinQ is building the networking fabric for quantum machines, the same role Ethernet and TCP/IP played for classical computers decades ago. ### Product and Technology The company builds high-finesse microcavities with micromirrors that create photonic links between isolated quantum processors. These cavity-enhanced connections unify separate processors into a single distributed system. The design is platform-agnostic, meaning it can integrate with different quantum processor architectures rather than locking into one hardware vendor. That flexibility broadens the addressable market compared to hardware-specific approaches from IonQ or IBM. ### Use of Proceeds and Vision The seed will fund a specialized lab in Cambridge, team growth, and early prototype demonstrations. CavilinQ wants to become the default interconnect layer for quantum computing within five years. If quantum scales the way the industry expects, someone has to build the plumbing. CavilinQ is betting it can be them. ### Market Context The quantum computing market is projected at $4.24B by 2030, with quantum networking growing from $1.15B in 2025 to $42.11B by 2035 at a 43.4% CAGR. IonQ, Photonic Inc., IBM, and Cisco are all working on interconnect solutions, but CavilinQ’s platform-agnostic cavity approach and its founding team’s research depth give it a distinct position in the race. ## 3. Miravoice Raises $6.3M to Replace Call Centers with AI Voice Agents ### Deal Overview - **Stage:** Seed - **Sector:** AI / Market Research - **Geography:** California, United States - **Round Size:** $6.3M - **Investors:** Unusual Ventures, Neo, 25madison ### Investor Profile Unusual Ventures led the round, joined by Neo and 25madison. Angel investors from Ramp, PubMatic, Atlassian, and Google also participated. The angel list suggests the founders’ network runs deep in enterprise SaaS, which matters for a product that sells into research firms and large organizations. ### Company and Leadership [Miravoice](https://www.miravoice.ai) was founded by Nishant Jain (CEO), Danny D. Leybzon (CTO), and Shreyas Tirumala (COO), three friends from California who’ve known each other for over a decade. The idea came from direct experience with the pain of scaling survey research. This is their first disclosed round. ### Problem and Opportunity Phone surveys still matter. Polling firms, market researchers, and universities need structured data from real conversations. But hiring, training, and managing human interviewers is expensive and slow. Response rates keep dropping. Interviewer costs keep rising. The $96.77B market research industry needs a way to conduct rigorous phone surveys at a fraction of the cost, without sacrificing methodological quality. ### Product and Technology Miravoice builds AI voice agents designed for long-form quantitative surveys. These aren’t chatbots. They handle 120+ question surveys lasting 40+ minutes, covering open-ended responses, Likert scales, matrix questions, and skip logic. A no-code interface lets non-technical teams upload questionnaires and launch campaigns. The company has already completed hundreds of thousands of production calls, with 10-20 customers in paid production. General-purpose voice AI tools like Bland AI or Vapi can’t maintain the methodological rigor that quantitative research demands. ### Use of Proceeds and Vision The funding goes toward expanding platform capabilities and growing the customer base across market research firms, polling organizations, and university departments. Miravoice positions itself as the replacement for CATI call centers, making phone-based data collection accessible and affordable at scales that were previously impractical. ### Market Context The global market research services market sits at roughly $96.77B in 2026. The public opinion and polling market alone is $3.63B, projected to hit $5.7B by 2034. Miravoice competes against traditional CATI providers like Dynata and Ipsos, general voice AI platforms, and survey software like Qualtrics. Its edge is domain-specific rigor that general tools don’t offer. ## 4. Anvil Robotics Raises $5.5M to Sell Legos for Robots ### Deal Overview - **Stage:** Seed - **Sector:** Deep Tech / Robotics - **Geography:** San Francisco, United States - **Round Size:** $5.5M - **Investors:** Matter Venture Partners (lead), DNX Ventures, Spacecadet Ventures, Humba Ventures, Position Ventures ### Investor Profile Matter Venture Partners led both the $1M pre-seed and this $5.5M seed, doubling down after seeing early traction. DNX Ventures brings cross-border expertise between the US and Japan, relevant for a company manufacturing in Taiwan. The five-firm syndicate reflects broad interest in the physical AI infrastructure thesis. ### Company and Leadership [Anvil Robotics](https://anvil.bot) was founded in July 2024 by Mike Xia (CEO) and Vijay Pradeep (CTO). Pradeep co-founded hiDOF, a robotics consulting firm, and led sensor calibration in Google’s AR/VR division. The company already has 50+ customers, has shipped 100+ units, and hit seven-figure revenue within roughly eight months of operation. ### Problem and Opportunity Physical AI teams building robots typically spend six months or more sourcing arms, cameras, and software libraries just to get a prototype working. The hardware side of robotics is fragmented and slow. While billions flow into AI foundation models for robotics, the teams building those models still struggle to get reliable, configurable hardware quickly and cheaply. ### Product and Technology Anvil sells modular robot devkits, hardware blocks that teams can customize and assemble into different configurations. Robots cost between $1,900 and $10,000 and ship within one to two days via air freight from Taiwan. Anvil designs and manufactures everything itself, a vertical integration play that gives it cost and speed advantages. Customers include Nvidia’s GEAR lab, Google, Co.bot, and Path Robotics. ### Use of Proceeds and Vision The seed will scale manufacturing capacity, expand the modular hardware catalog, and grow the customer base. Anvil wants to be the default hardware platform for physical AI, the way AWS became the default cloud for software startups. If every team building embodied AI needs configurable robots fast, Anvil is positioning to be the supplier. ### Market Context The AI robotics market is valued at $20.4B in 2025, projected to reach $182.7B by 2033 at a 32% CAGR. Q1 2026 alone saw over $2.26B in robotics funding. Anvil competes less with NVIDIA Isaac or Google DeepMind and more with the status quo: teams cobbling together hardware from multiple suppliers over months. Its hardware-first, modular approach fills a gap that software-focused competitors don’t address. ## Lessons For Founders - **Operator founders attract outsized conviction.** Whirl AI’s Sunny Bedi (20 years as a CIO) and Anvil’s Vijay Pradeep (Google AR/VR, robotics consulting) brought direct experience with the problems they’re solving. ICONIQ doesn’t lead seed rounds, but they did for Bedi. Build something you’ve lived through. - **Narrow beats general at seed stage.** Miravoice doesn’t do general voice AI. CavilinQ doesn’t do general quantum hardware. Whirl doesn’t do general enterprise search. Each company picked a specific wedge where general-purpose tools fall short, and that focus is what makes the thesis investable early. - **Revenue before seed is the new normal.** Anvil hit seven-figure revenue in eight months. Miravoice has 10-20 paying customers and hundreds of thousands of production calls. Investors at seed now expect working product and real customers, not just prototypes and TAM slides. - **Platform-agnostic infrastructure plays well in emerging categories.** CavilinQ’s interconnects work across quantum architectures. Anvil’s hardware works across AI software stacks. When a category is early, picking a side is risky. Building the layer that works with everyone is a safer bet.