No Creds Notes #23
Qubits, Drones, and Blood Pressure
Hey everyone!
Hope the week’s treating you well. We’ve got three good ones this week, quantum computing proving itself in materials science, a startup winning half a billion dollars from the Pentagon, and an awesome new blood pressure drug. Let’s get into it.
If this is your first time reading Uncredentialed, welcome! Subscribe below to receive regular updates on tech, startups, and strategy.
The Qubit Gets a Job
Back in the early days of Uncredentialed, I covered Google’s exciting announcement that researchers figured out how to verify quantum calculations by running a process forward and then backward, catching the “echoes” of real quantum dynamics. It was a critical step: you can’t build anything reliable on a machine you can’t verify.
This week, Q-CTRL took the next step by actually using a verified quantum machine to do something useful.
On May 6th, the Sydney-based quantum software company published results showing a 3,000x speedup on a materials science simulation that runs on IBM’s quantum hardware. The same calculation took classical computers over 100 hours. The quantum run finished in 2 minutes.
The problem they solved is called the Fermi-Hubbard model, a standard benchmark in condensed matter physics that simulates how electrons interact in materials, allowing researchers to design better battery cathode materials, understand high-temperature superconductors, or engineer new industrial catalysts. Right now, roughly one-third of all global supercomputer time goes to chemistry and materials simulations like this one, just waiting to be disrupted by quantum.
The improvement came from Q-CTRL’s Fire Opal software, a layer that sits on top of IBM’s hardware and dynamically recalibrates gate operations to suppress decoherence noise (this is the main thing that makes quantum calculations go wrong, basically the qubit somehow interacts with its environment, causing it to lose its simultaneous 1,0 superposition state and turn into classical data). Fire Opal has proven itself to be able to push qubits well beyond the baseline fidelity of IBM’s native toolset.
To be clear, this did not involve innovation at the hardware layer (IBM’s qubits remained the same). The software layer is what turned a fast-but-noisy machine into something that could outperform classical hardware on a commercially relevant problem. We can think of it sort of like how of course GPUs were valuable even before AI, but the invention of transformers and everything since has simultaneously made both the software and hardware layer go exponential in value.
If the result generalizes, the implications are significant. As I mentioned earlier, materials simulation drives the design of the next generation of batteries, catalysts, and semiconductors. Classical supercomputers are throwing enormous resources at problems that, it now appears, quantum machines may be able to solve orders of magnitude faster. A world where this revolution in materials science occurs would have phones and laptops that go weeks between charges, wearables like Whoop or Oura being replaced by “smart” clothes that track your metrics just as well while feeling the same as any old t-shirt, EVs that can drive cross country on a single charge, lower energy bills due to a better conducting electrical grid, lower food prices because of enzymatically-created fertilizers, the end of plastic pollution as we know it, and much much more.
Breaking The Valley of Death
Prior to the outbreak of the Ukraine war, you would’ve been the Rockwell Freedom of Speech guy meme (see below) alongside Noah Smith if you were emphasizing the role drones would play in future warfare.
A few short years later and every major conflict is being carried out first and foremost via drone warfare. Meanwhile the US has effectively no drone or drone defense production.
On May 19, Perennial Autonomy signed up to help fix that problem. The startup signed a $500 million contract with JIATF-401, the Pentagon task force set up specifically to buy counter-drone technology. The deal covers Merops interceptors, Bumblebee quadcopters, and Hornet midrange strike drones. Army Secretary Daniel Driscoll said Merops production is already scaling.
It’s particularly exciting because aside from Anduril and a few select others, the defense tech startup space has been somewhat of a valley of death. Build something impressive, raise venture dollars, win a pilot, get stuck in procurement forever, then watch the big primes absorb the real money. Instead, Perennial won the $500M contract and are still nimble enough to excel in a vertical like drones where speed matters more than polish. Counter-drone is basically an iteration war. One side changes the drone, the other side changes the detector, jammer, or interceptor, then the first side changes the drone again, month after month after month. That’s a terrible fit for the classic defense acquisition machine, which was built for aircraft carriers, missile programs, and platforms planned years in advance.
That mismatch is why JIATF-401 exists. The Pentagon is effectively admitting that some threats cannot be bought through the old tempo. If adversarial drones are updating at commercial hardware speed, the buyer has to move closer to commercial hardware speed too. This gives startups an opening to nudge their way into the space and, hopefully, get in enough to instill widespread systemic modernization and efficiency gains.
Before I get too carried away opining on the broken state of our military-industrial supply chain, let’s talk about Perennial’s drones. Merops is an AI-enabled interceptor meant to find and hit drones in environments where a human-in-the-loop may simply be too slow. If you’re keeping up with the Ukraine or Iran conflicts, the drones these seek out are typically the Shahed/Gerbera style drone that have wreaked so much harm through recent conflicts. Meanwhile, the Bumblebee is a smaller, quadcopter-style drone intended to autonomously target and take out opposing quadcopter drones. Lastly, the Hornet is an offensive drone used in a similar kamikaze style to the Shahed drones, but with much better autonomy and precision.
Perennial still has plenty to prove. A contract is not the same thing as battlefield dominance, and scaling defense hardware is where startup dreams go to get humbled, but with the rising importance of drone warfare, it’s good to see us stepping up on the investment and innovation front.
Going to the Source
I realize a hypertension drug is not as sci-fi as quantum computers or autonomous interceptors, but I seem to be making it a habit to include exciting medical/drug progress in these posts.
This week, AstraZeneca announced the FDA approval of Baxfendy (baxdrostat), its new treatment for uncontrolled hypertension and the first approved drug in its class. Hypertension affects roughly half of American adults, and AstraZeneca says about 23M US patients remain uncontrolled despite taking 2 or more blood pressure medicines. In other words, “just take blood pressure medicine” does not quite solve the problem for a very large group of people.
Most current blood pressure drugs work downstream. ACE inhibitors, ARBs, beta blockers, calcium channel blockers, and diuretics all use different tools and different pathways, but the common pattern is that they intervene after the body has already started the hormonal cascade that raises blood pressure.
Conceptually, these traditional medications are just trying to manage a flood after the levy has broken. They fall into 2 reactive camps:
The Hardware Dampeners (Beta Blockers & Calcium Channel Blockers): These ignore the hormonal software entirely and act as mechanical shock absorbers on the physical plumbing by either forcing the arterial walls to relax or putting sort of muffler on the heart, preventing it from revving its engine in response to adrenaline.
The Mid-Stream Interceptors (ACE Inhibitors, ARBs, & Diuretics): These try to wrestle with the hormonal loop mid-flight. ACE inhibitors and ARBs try to block or blindfold a messenger hormone called Angiotensin II. If that fails, the diuretics wait at the very end of the line, forcing the kidneys to flush out excess fluid after the hormonal alarm has already caused the body to hoard it.
The fatal flaw in this entire traditional lineup is a clinical phenomenon known as aldosterone escape. Because the body is highly redundant, it eventually routes around these mid-stream blocks, finding alternative pathways to keep churning out the final, heavy-hitting hormone that commands your kidneys to hold onto salt and water. Once that hormone breaks through, blood pressure spikes right back up.
Baxdrostat goes upstream. It blocks aldosterone synthase, the enzyme that produces aldosterone. Aldosterone tells your kidneys to hold onto sodium and water, which raises blood pressure. Less aldosterone, less sodium/water retention, lower pressure.
By going straight to the source inside the adrenal gland and shutting down the factory worker responsible for building the actual problem-causing hormone, Baxdrostat bypasses the body’s redundancy loops entirely.
Aldosterone synthase is encoded by something called CYP11B2, while cortisol synthase is encoded by CYP11B1. Because these two enzymes sit right next to each other in the adrenal gland and share a nearly identical genetic sequence, early drug candidates couldn’t tell them apart, hitting both and dangerously suppressing cortisol production in the process. Triggering adrenal insufficiency just to lower blood pressure is a terrible trade.
Baxdrostat finally cracked this structural puzzle by achieving a massive 100x selectivity for CYP11B2 over CYP11B1. That margin of safety transforms it from a great pharmacological concept into a viable, FDA-approved therapeutic. Clinical trial data showed that adding Baxdrostat on top of patients’ existing regimens triggered a 9.8 mmHg drop in systolic blood pressure. In cardiovascular medicine, a double-digit reduction is a profound victory, translating directly to a significantly lower baseline risk for stroke, heart attacks, and major cardiovascular events.
To ground the scale of that impact, a reduction of this magnitude across a high-risk population could prevent more than 100,000 strokes and heart attacks every single year in the United States alone.
If you enjoyed this post consider sharing it with a friend! And if you haven’t already, subscribe below!