The Check Is Not the Door

Niron Magnetics

The Shakopee Mdewakanton Sioux Community did not arrive at Niron Magnetics when a factory made the company obvious. The sovereign Dakota tribal government first backed Niron in 2016, when rare-earth-free Iron Nitride was still fighting its way from university research toward repeatable manufacturing.

Now SMSC is lending Niron $150M to help build that process at full scale.

The financing will support construction, equipment, and operational readiness at Niron's Sartell, Minnesota plant. Niron expects production to begin in 2027, with capacity of up to 1,500 tons of permanent magnets annually and about 175 full-time jobs.

This is separate from the Office of Strategic Capital's earlier conditional commitment for a direct loan of up to $150M. The public capital stack is getting larger, but the SMSC loan's rate, maturity, security, and covenants were not disclosed.

Niron is trying to remove rare-earth dependency from the material itself. Co-founder Jian-Ping Wang's University of Minnesota research produced magnets from iron and nitrogen. CEO Jonathan Rowntree and CTO Frank Johnson now have to turn that breakthrough into yield, quality, customer qualification, and reliable output.

That is where SMSC's role gets interesting. The Community was one of Niron's earliest investors, joined the company's $33M 2023 round, returned for its $25M 2024 strategic financing, and is now crossing from equity into lending.

This is not a lender meeting a project through a spreadsheet. It is a partner increasing its exposure after watching the technology, team, and manufacturing plan develop for a decade.

Chairman Cole Miller connected that decision to Dakota stewardship and Minnesota economic opportunity. Shared values can explain why a relationship began, but a $150M loan still has to meet construction schedules, equipment milestones, production economics, and customer demand.

Niron plans to put material-to-magnet production under one roof, then use Sartell as the first module in a larger U.S. platform. A proposed second facility is designed for up to 10,000 tons annually, with groundbreaking targeted for 2028.

The ambition is national. The risk is operational. Laboratory performance does not automatically survive procurement, installation, commissioning, scrap rates, workforce training, quality systems, and volume.

SMSC saw Niron before the factory. The loan says it still sees the company after the work became expensive, physical, and unforgiving. Rare-earth independence will not be won by announcing a better material. It will be won by producing it, repeatedly, when customers need it.

Dirac Labs

The hardware at the center of Dirac Labs starts with a flaw in a diamond. That engineered nitrogen-vacancy center can detect tiny variations in Earth’s magnetic field, giving a vehicle another way to understand where it is when a satellite signal disappears.

Dirac Labs has raised $1.8M in pre-seed financing to turn that physics into equipment that can survive outside a research setting. TitletownTech led the round, with Automotive Ventures, Riceberg Ventures, quantumEDGE Ventures, Jude Gomila, and Balaji Srinivasan participating.

The money will fund prototypes, field trials, and further refinement of the company’s universal positioning system. Co-founder and CEO Sanket Deshpande and co-founder and COO Aishwarya Das are building diamond-based quantum magnetometers paired with AI signal processing and sensor fusion for navigation underwater, underground, in remote terrain, and across contested environments where GPS may be unavailable, jammed, or spoofed.

The sensor reads local magnetic signatures. Software compares them with magnetic maps and processes the signal onboard. Dirac says the architecture can connect through existing GPS device ports on aircraft, submarines, and other vehicles rather than demanding a platform rebuild.

Quantum sensing has spent years proving extraordinary measurements are possible in controlled environments. Customers in defense, maritime, mining, agriculture, and autonomy care about vibration, temperature, noise, manufacturing variation, integration, and procurement. A sensor that is brilliant on a bench still has to earn its place where a bad position fix becomes an operating event.

Dirac told EE Times it is working with unnamed U.S. defense primes and agritech companies as integration partners. The company says its current sensor package is roughly 5 centimeters across, with a longer-term plan for semiconductor-compatible manufacturing. Customer names, contract values, revenue, production economics, and deployment volume remain undisclosed.

That distinction is where this pre-seed round earns its weight. $1.8M is not buying a victory lap for quantum navigation. It is buying field time, manufacturing work, integration evidence, and the chance to discover which parts of the system stop behaving politely when the laboratory door opens.

Dirac Labs is working where GPS infrastructure reaches its limits. The next proof will not be that a diamond can sense Earth’s field. It will be whether the whole system can hold a vehicle’s trust when the map goes quiet.

Shape Memory Medical

Shape Memory Medical raised $10M in convertible-note financing for the part of medical technology nobody can accelerate with a better headline.

Follow-up.

August Global Partners led the financing. Taiwania Capital joined as a new investor, with support from HEAL Venture Lab.

The structure matters. This is a convertible note, not a priced equity round. Shape Memory Medical gets capital now while some of the ownership economics wait for a future conversion event.

Ted Ruppel and the team are using that time across two very different clinical clocks.

The first is AAA-SHAPE, a randomized pivotal trial evaluating the company's shape-memory polymer technology alongside endovascular repair for abdominal aortic aneurysms. Enrollment was completed in July with 180 patients across 48 centers in the U.S., Europe, and New Zealand. Follow-up is planned for five years.

The second is FLAGSHIP, an earlier first-in-human feasibility study in aortic dissection.

Shape Memory Medical's material is compressed for catheter delivery and designed to expand after contact with blood. The company reports use in more than 3,500 patients and approvals in more than 40 countries.

Those figures show a platform with operating history.

They do not finish either of these evidence programs.

Completing enrollment is not a clinical outcome. A material working in one vascular setting does not guarantee that it travels into another. The company still has to retain patients, protect data quality, analyze the results, and make a case clinicians and regulators can trust.

That is what the note is bridging.

The polymer can change shape inside the body in minutes. The evidence around it still has to accumulate one patient, one visit, and one unanswered question at a time.

EULER

The most interesting part of EULER's $4.3M Seed is that the company did not need the money to prove somebody would pay.

It had already bootstrapped to profitability.

Channel Equity Partners led EULER's first institutional round after the company says revenue grew 600% in 2025 and doubled again during the first half of 2026.

Treat those as company-reported figures, but do not miss what the sequence changes.

Greg Portnoy is not taking venture capital to rescue the business from a missing market. He is taking it to move faster than current revenue would otherwise allow.

EULER is building an agentic partner relationship management platform around work that usually disappears into portals, spreadsheets, inboxes, and the heads of a few people who know which relationship matters.

The company says the system supports more than 100,000 partnerships. Since June, it reports answering 50,000 questions for 7,700 partners and saving an estimated 31,000 hours. Customers include Gong, BambooHR, Zip, and BigCommerce.

That is usage. The next proof is judgment.

Partner programs rarely fail because nobody could find the enablement deck. They fail because incentives drift, ownership gets fuzzy, context disappears, and the relationship slowly becomes a ticket queue.

EULER is betting more of that context can become accessible and actionable without turning the partner into another person receiving a faster automated answer.

The Seed will support hiring, product velocity, and go-to-market work. The company plans to double its team within six months.

Bootstrapping forced EULER to build at the speed of customer revenue.

Outside capital gives it permission to build ahead of that revenue, which is useful right up until speed begins deciding what the product is allowed to forget.

PayQuicker

PayQuicker has spent eighteen years making the final mile of a payment look simpler than the infrastructure underneath it.

Clearhaven Partners just made a significant strategic investment in the company. The amount and ownership terms were not disclosed.

Paul Beldham and other members of PayQuicker's leadership team invested alongside Clearhaven.

That is the part I would keep in the frame.

PayQuicker is not an early startup raising around a payments diagram. Founded in 2008, it says one API now connects businesses to banks, card networks, processors, and local partners across more than 210 countries and territories and more than 80 currencies.

Every payout still carries a set of decisions the recipient never sees.

Cost. Speed. Currency. Method. Reliability. Compliance. Which rail is working in which market at that particular moment.

The product has to hide enough of that complexity to feel useful while preserving enough visibility to explain what happened when the money does not arrive as expected.

Clearhaven manages more than $1B and focuses on software. The firm says the investment will support network expansion, geographic reach, commercial capabilities, adoption, and customer experience.

Because the terms are private, there is no responsible way to assign a valuation, invent a round size, or assume this is a control transaction.

What we can see is alignment.

Management put capital beside the investor, tying the people operating the network to the ownership case being built around it.

More rails will widen PayQuicker's reach. They will also widen the compliance perimeter and the number of relationships that can fail outside the company's direct control.

The investment is a bet that a mature network can become a larger platform.

The next proof lives inside the payments nobody notices because they arrived exactly when, where, and how they were supposed to.

Quintessent

Quintessent announced a $40M Series A on the same day it began handing its product to customers.

The second event will tell us considerably more about the first.

Cycle Capital led the oversubscribed round. New investors included Goldman Sachs XIG-Industry Ventures, Hina Liberty Capital, Susquehanna International Group, InterVest, Safar Partners, and Ciena. Foothill Ventures, M Ventures, Osage University Partners, and Sierra Ventures returned.

Alan Liu and the Quintessent team are building a quantum-dot comb laser designed to generate eight optical wavelengths from one chip with one bias control.

That matters because optical systems usually pay for capacity with more components, more controls, more power, and more places for integration to become expensive.

Quintessent's argument is that a gallium-arsenide quantum-dot source integrated with silicon photonics can simplify that architecture while reducing reliance on a constrained indium-phosphide supply chain.

The company says the design can reduce component count, power, and complexity.

Now somebody else gets to measure it.

Customer sampling moves the product out of the development narrative and into qualification environments Quintessent does not control. Prospective customers can test optical performance, temperature stability, packaging, manufacturing variation, reliability, and whether eight wavelengths on one chip create enough system-level value to change an existing design.

Ciena's place on the cap table adds an important industry perspective. It does not establish a customer order or a design win.

The Series A gives Quintessent more capacity to become a product company. The evaluation kit starts determining whether the product can become part of someone else's infrastructure.

The round measures investor conviction.

The first boxes coming back from customer labs will begin measuring what happened after the light left Quintessent's chip.

PayRewards

Pay.com.au has raised US$28M to find out whether an Australian payments habit can become an American one.

The Series E will fund the U.S. launch of PayRewards. Existing and new investors participated, but the company did not identify them. Pay.com.au says the round brings total capital raised to US$70M.

The parent already has an operating record.

It reports processing US$7B during the last twelve months across 30,000 Australian businesses, with 100% year-over-year growth. Those are company-reported figures, but they explain why the next market is being attempted now.

PayRewards lets U.S. businesses earn points on payments that normally feel like pure operating expense, including ACH transfers, bank payments, and card transactions. Its Double-Dip feature is designed to layer PayRewards points on top of an eligible card's existing rewards.

There is no monthly platform fee. The business pays when it chooses to earn.

That sounds like a product feature until the economics start moving underneath it.

Rewards have to be valuable enough to change payment behavior. Processing costs, partner economics, fraud controls, cash-flow timing, and customer acquisition still have to leave a business behind.

Blake Hutchison is leading the U.S. operation. The service is available in most states, not all of them, which makes this a rollout still being built rather than a nationwide victory already completed.

The capital is not financing the discovery of the original model. Pay.com.au has already produced evidence in Australia.

It is financing geographic translation.

The payment rails are different. The regulatory map is different. The rewards market is louder. The American SMB already has a wallet full of products asking to sit between the business and its bills.

The $28M gets PayRewards into that competition.

The first meaningful signal may be which expenses a U.S. owner is willing to reroute before the points begin changing what happens after the payment clears.

Thea Energy

Fusion capital is beginning to move past the question of whether the science is interesting.

The harder question is whether the machine can be built repeatedly, financed sensibly, and turned into power somebody can actually buy.

Thea Energy has extended the $100M Series B it announced in May. The company did not disclose the size of the extension, so this should not be read as another $100M round.

New support came from Brevan Howard Macro Venture, Aloniq, ALJ Investments, Beyond Earth Ventures, and additional strategic investors.

The timing tells the better story.

Since the original Series B, Thea Energy has reported a separate $20M ARPA-E SCALEUP award, magnet validation work, a digital-twin collaboration with NVIDIA, Synopsys, Argonne National Laboratory, and Princeton Plasma Physics Laboratory, and progress evaluating sites and speaking with potential power customers.

Those are different forms of capital doing different jobs.

The equity supports the company. The federal award supports a defined technical program. Strategic investors may add industrial reach. None of them makes the engineering risk disappear.

Brian Berzin and the team are now using the added capacity to scale magnet manufacturing, open a second Northern New Jersey facility, build the Eos integrated fusion system, and move the proposed Helios plants closer to deployment.

Thea's argument is that stellarator fusion can keep its plasma-stability advantages while replacing some of the most difficult magnet geometry with a more manufacturable planar system controlled through software.

That thesis has attracted serious capital.

Eos is where it has to become a serious machine.

The extension is the handoff between the two, and the amount left undisclosed may end up being less revealing than what the next piece of hardware can hold.

Mafix

Carbon removal gets considerably more interesting when somebody else has a reason to buy the product first.

Mafix has raised a $5.4M pre-seed led by Azolla Ventures, with Counteract VC, Astera Institute, Plug and Play Ventures, Impact Science Ventures, and an undisclosed Dutch family office participating.

The company is building a silicon fertilizer with two jobs.

The first is agricultural. Silicon can help crops manage stress, but naturally occurring silicate minerals often weather too slowly to deliver plant-available silicon on a useful farming timeline.

The second is atmospheric. Speeding up that weathering process can also remove carbon dioxide.

Jade Marcus and Stanford chemistry professor Matthew Kanan are trying to make those two economics travel together.

That is the part worth watching.

Most carbon-removal companies eventually have to answer who pays before the carbon market becomes large, liquid, and dependable enough to carry the business. Mafix wants the fertilizer to create value in the field while the same material creates a measurable carbon record.

The company says its process can convert silicate rock into a product that weathers within one growing season and works with existing farm equipment. It also plans to use spare cement-kiln capacity rather than build a new first-of-kind plant.

The new capital will fund larger field trials and a commercial demonstration expected to produce at least 1,000 tons.

That is where the two stories have to meet.

Farmers will care about yield, handling, consistency, and price. Carbon buyers will care about measurement, permanence, and whether verification leaves enough economics behind. Contract manufacturing adds another handoff inside someone else's industrial environment.

The pre-seed does not prove that fertilizer can finance carbon removal.

It gives Mafix enough material, enough acreage, and enough operating reality to find out whether one customer need can carry the other further than climate enthusiasm alone.

Cityblock Health

Cityblock Health did not raise $116M simply to become a larger version of itself.

It raised the Series E while signing an all-stock agreement to acquire Homeward, extending a care model built around Medicaid and dual-eligible populations into rural Medicare Advantage.

General Catalyst led the financing. Cityblock disclosed the size of the round. It did not disclose a purchase price for Homeward beyond describing the transaction as all-stock.

Keep those two events separate and the operating question becomes clearer.

Cityblock says it serves almost 200,000 members and is generating $2.2B in annualized revenue, up 77% year over year. Homeward reports nearly 50,000 members through a rural-first model.

Those figures establish scale. They do not complete the integration.

Toyin Ajayi and the Cityblock team have spent nearly a decade building around patients whose medical needs often sit beside housing, transportation, food access, behavioral health, and the rest of a life that refuses to fit inside a clean clinical workflow.

Jenny Schneider, Amar Kendale, and Homeward built for a different operating environment, where distance, workforce constraints, provider access, and local trust change how care reaches the patient.

Cityblock's CORE platform may help connect the two. The company says its data and automation have freed 44,599 clinician hours and handled 42,148 calls through AI in 2026. Those are company-reported operating metrics, not proof that the system will transfer cleanly into every rural community.

That is the work the capital has bought.

Not simply more members. A larger responsibility.

The combined company has to show that technology can make care teams more prepared without making care feel more distant, and that scale can survive the reimbursement pressure waiting on the other side of growth.

The Series E creates capacity. The all-stock deal chooses a direction.

Somewhere between the two, Cityblock has to prove that a high-touch care model can travel without leaving the touch behind.