Barron Stone

I'm starting a new movement...Acquisition Accelerationism (a/acc) emulating the tenants of e/acc: Acceleration as Imperative - Traditional acquisition timelines (PALT, post-JCIDS, milestone reviews) are mismatched with technological velocity. - a/acc asserts that compressing the timeline from idea → contract → capability delivery is not optional but existential for national security. AI-First Acquisition - Every acquisition function—requirements generation, source selection, cost estimation, contract writing, program oversight—should be re-imagined with AI as a co-pilot. - AI is not an add-on, but a paradigm shift that transforms the throughput, accuracy, and adaptability of acquisition processes. Experimentation > Bureaucracy - a/acc values rapid pilots, prototyping, and data-driven feedback loops over endless documentation. - Just as Silicon Valley uses MVPs, acquisition should use “minimum viable solicitations” and “minimum viable contracts” to accelerate learning. Decentralized Dynamism - Power should shift from central gatekeepers to empowered program teams with AI-enabled toolchains. - Competition between agencies, consortia, and small businesses creates evolutionary pressure that accelerates innovation. Human-Machine Synergy - Bureaucrats, program managers, and contracting officers become augmented operators, not paper-pushers. - a/acc reframes the workforce not as bottlenecks but as AI-empowered catalysts for industrial base expansion. Moral Duty of Speed - Delays in adopting new tech are framed as ethical failures. - Every year lost in procurement slowness is a year the warfighter lacks lifesaving capabilities, and a year adversaries can surge ahead. Civilizational Security & Abundance - a/acc emphasizes that accelerating acquisition is not just about faster weapons, but about sustaining U.S. technological leadership, industrial base vitality, and global stability. - Technology adoption in acquisition becomes a civilizational defense strategy. I'm making stickers and t-shirts... Join the movement...

69

26 Comments

“Building your own system will cost less and work better.” That’s the biggest lie in aerospace and manufacturing ops. I hear it all the time: “We’ll just hack together something in Excel.” “Our engineers can spin up a tool for free.” “Paper’s fine, we’ve always done it this way.” Here’s the reality: * DIY tools are never cheap. By the time you factor in upkeep, custom patches, and the one engineer who becomes the bottleneck for every request, you’ve sunk years of time and millions in hidden costs. * Paper isn’t “simple.” It’s a risk. Missed steps, bad handwriting, missing binders, small mistakes that turn into schedule slips, blown tests, or safety issues. * Internal tools don’t scale. What works for 20 people collapses at 200. Then you’re rebuilding under pressure, mid-contract, when failure isn’t an option. Companies in space, defense, energy, robotics, and other high-stakes industries don’t lose because they’re less innovative. They lose because their processes fail them, missed traceability, compliance gaps, and siloed teams. The real risk isn’t adopting a new system. The real risk is sticking with the one you think is “good enough.” The better path: * Standardize procedures across the org * Build in traceability and accountability from day one * Use tools designed by teams who live and breathe complex ops, not “side projects” from internal devs

40

8 Comments

Every defense manufacturer I talk to is facing the same problem: demand is surging and they can't hire fast enough to keep up. The contracts are there. The budgets are approved. But production is bottlenecked by one thing—getting enough trained people on the floor who can build complex assemblies without introducing defects. The training problem nobody wants to talk about, traditional manufacturing training is slow. Really slow. → Experienced technicians spend weeks shadowing before they're trusted on critical assemblies → Tribal knowledge lives in the heads of your most senior people—and they're retiring taking it out the door. → Documentation is often outdated, incomplete, or ignored → Every mistake during the learning curve costs time, material, and rework When your product complexity is increasing and your experienced workforce is shrinking, this math gets ugly fast. The manufacturers scaling successfully in 2026 have flipped the training model: ✅ Video work instructions that actually get used — Step-by-step visual guidance at the station. No more binders collecting dust. Workers watch, do, and get real-time verification they did it right. ✅ Novice to productive in hours, not weeks — We have customers train novice workers to complete a precision assembly process in 20-30 minutes. 10x faster. That same process used to take hours with their most skilled people. ✅ Knowledge that transfers instantly — When you improve a process at one site, it replicates everywhere immediately. No retraining tours. No version control nightmares. ✅ Real-time error prevention — The system catches mistakes as they happen, not at final inspection. New workers build confidence because they know they're doing it right. ✅ Language and literacy barriers disappear — Video transcends both. Suddenly your hiring pool gets a lot bigger. 🎯 Our customers at Rapta, Inc. are seeing 30%+ capacity gains without expanding headcount. Not because they're working people harder—because they're making every worker more capable from day one.

19

In #defensemanufacturing, speed isn’t a metric, it’s survival. Firestorm Labs was born to build faster, smarter, and closer to the fight. Founded in 2022 in San Diego by Dan Magy 🏴‍☠️, Ian Muceus, and Chad McCoy, the company rose from a #wartime reality check: if missions shift in hours, why does production crawl for quarters? The answer became a revolution, manufacturing sovereignty at mission tempo. Firestorm’s edge is modularity with intent. Its flagship UAS, Tempest, isn’t a drone, it’s a platform. Wingspan: 7–9.5 ft. Payload: 10 lbs. Reconfigurable in minutes. Portable enough for one operator to break down and move. The quick-connect design makes even the airframe modular, turning every flight into a plug-and-play operation. Tempest flies in the Group 2/3 class but costs one-fifth of traditional systems and builds in hours, not weeks. In defense, that isn’t iteration, it’s disruption. Under the shell runs OCTRA, “One Chip to Rule Them All,” Firestorm’s proprietary flight controller, compute core, and power brain. It enables #AIautonomy, GPS-denied navigation, and open-architecture integration that scales across missions. Pair it with xCell, the expeditionary factory-in-a-box, and Firestorm goes off-grid: two containers, 50 units per month, full production capability powered by a generator. This isn’t defense supply, it’s defense on demand. Investors moved fast. In July 2025, Firestorm closed a $47M Series A led by New Enterprise Associates (NEA) with Lockheed Martin Ventures, Booz Allen Hamilton Ventures, Decisive Point, and Washington Harbour Partners LP. Add $12M venture debt from J.P. Morgan and a $100M United States Air Force IDIQ contract running through 2031, and you get altitude and authorization. Total raise: $59.5M, capital chasing capability at full throttle. The founders bring three distinct battlefields of experience. CEO Magy built and sold Citadel Defense in a nine-figure deal. CTO Muceus engineered #aerospacemanufacturing breakthroughs at Origin and Stratasys. CSO McCoy logged 17 combat deployments with JSOC’s 24th STS as a Pararescueman. Together, they form a feedback loop of vision, precision, and execution that turns defense from #procurement into performance. Firestorm calls it “democratizing deterrence,” building tech where timelines match threats. With roughly 80 employees across CA and FL and a new 35K sq. ft. facility at MCAS Miramar, the company is scaling like a unit on alert. They’re hiring across #engineering, manufacturing, and #operations because in this game, talent equals ammunition. #Startups #StartupSpotlight #StartupJourney #StartupGrowth #FounderStory #AI #Manufacturing #ManufacturingTech #Defense #DefenseTech #DeepTech #Disruption #Technology #Innovation #TechEcosystem #StartupEcosystem #Hiring #TechHiring If software engineering peace of mind is what you crave, Vention is your zen.

21

3 Comments

Iterative Systems Engineering: In Practice is here! We heard your feedback. Teams wanted something they could share with peers and put into practice together. A shared playbook on how to design, test, and iterate complex systems. This 40-page tactical guide is built for the team doing the work. It’s grounded in how top hardware teams actually run fast, iterative V-cycles with testable milestones, evolving architectures, and continuous verification from day one. A clear, practical framework drawn from real programs. If you're a systems engineer, RE, or team lead looking for an agile framework that works, we wrote this for you. We're giving early access to 50 people ahead of the public release. 👉 Comment "handbook" below and we’ll send the PDF your way.

160

428 Comments

I’ve been getting back into hands-on mission design learning recently, revisiting astrodynamics, orbit determination, and the kind of tooling that actually scales beyond textbook problems. That’s when I stumbled upon Nyx Space and their open astrodynamics stack. At first glance, what stood out was the engineering discipline: • Computationally efficient libraries • Clean APIs • Designed for modern workflows, not legacy pipelines • Rust at the core, with sensible bindings for practical use As someone who has dealt with Monte Carlo runs, batch analysis, and concurrency pain points, the performance and thread-safety focus immediately made sense. This feels like software built for today’s constellation-scale thinking, not just single-mission analysis. Then I looked closer at the licensing. Nyx Space uses a strong copyleft model (AGPL-style), and that changed how I viewed the whole ecosystem. This isn’t just an open-source checkbox. Copyleft actively shapes behavior: • If you improve the core, those improvements flow back • Hosted or internal use doesn’t quietly bypass sharing • Fork-and-lock strategies become harder to justify • Collaboration becomes the default, not an afterthought In a field like space mission design, where correctness, edge cases, and domain knowledge accumulate slowly and painfully, this creates an interesting dynamic. Instead of fragmented internal tools and proprietary silos, the system nudges engineers toward shared foundations and upstream contribution. It’s a very different incentive structure compared to permissive licenses, and arguably closer to how we already collaborate informally across missions, papers, and reference models. Curious to hear from others: Have you worked with copyleft licensing before? How do you see its implications for aerospace and space software adoption? Could this kind of licensing model enable a more genuinely collaborative industry framework, or does it introduce friction for commercial programs? Would love to hear perspectives, especially from folks building or operating mission software today.

19

𝗕𝗼𝗮𝗿𝗱 𝘀𝗲𝗮𝘁𝘀 𝗮𝗿𝗲𝗻'𝘁 𝘄𝗼𝗻 𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝗮𝗽𝗽𝗹𝗶𝗰𝗮𝘁𝗶𝗼𝗻𝘀. 𝗧𝗵𝗲𝘆'𝗿𝗲 𝗲𝗮𝗿𝗻𝗲𝗱 𝘁𝗵𝗿𝗼𝘂𝗴𝗵 𝘁𝗿𝘂𝘀𝘁. Leon Rubinstein joined me on 𝘊𝘢𝘯𝘥𝘪𝘥 𝘊𝘰𝘧𝘧𝘦 to unpack the hidden reality of boards. The biggest misconception? Your résumé opens doors. In truth, boards select people they already trust to protect mission and reputation. If you want a seat, build credibility long before the invitation arrives. Watch the full episode on YouTube. Presented by Veblen Director Programme.

11

3 Comments

Manufacturers collect huge volumes of machine data, yet most of it stays unused because it lacks context. Merely wiring equipment to dashboards does not fix the deeper problems of integration, governance, and scalability. To turn raw signals into actionable intelligence, you need a unified approach. 𝐂𝐨𝐫𝐞 𝐜𝐡𝐚𝐥𝐥𝐞𝐧𝐠𝐞𝐬: ❌ Data silos that isolate critical information ❌ Complex integrations between legacy and modern systems ❌ No single source of truth ❌ Poor data quality feeding AI/ML models and dashboards 𝐓𝐡𝐞 𝐔𝐧𝐢𝐟𝐢𝐞𝐝 𝐍𝐚𝐦𝐞𝐬𝐩𝐚𝐜𝐞 𝐚𝐝𝐯𝐚𝐧𝐭𝐚𝐠𝐞: ✅ Centralized, real-time data architecture ✅ One authoritative “source of truth” spanning MES, SCADA, ERP, and cloud platforms ✅ A common data language so every machine and application can communicate seamlessly In his talk “Deploying Real-World UNS Architectures with MQTT and Node-RED,” at the recent CONNACK event hosted by HiveMQ, Alejandro Simó Vesperinas from Mayker outlined a three-step model: 1. Model and enrich data at the edge with Node-RED. 2. Publish structured data via MQTT. 3. Organize everything in a Unified Namespace ensuring all stakeholders, and algorithms, see the same, trusted information. Watch the full presentation below. https://lnkd.in/dJyT8_Gu

21

CadenceLIVE is May 7 and if you're leading Aerospace & Defense to the digital engineering future you need to be there. Regardless of your program constraints, you'll gain critical information you need to modernize. To help you maximize you're day, I've put together a handy field guide and I'm looking forward to seeing you there! https://ow.ly/Pqna50VNmkz #AerospaceDefense #DigitalEngineering #obsolescence #DMSMS #FPGA #ASIC #digitaltwin

23

Space is becoming a more contested domain—and we're stepping up to help Space Force maintain custody of what matters most. Vantor will support United States Space Force in tracking high-interest space objects through orbital blind spots, areas ground sensors can't reach. Our Non-Earth Imaging (NEI) capability delivers space-to-space imaging at <10 cm resolution from hundreds of kilometers away, giving Space Force critical intelligence about object position and trajectory when an object has the potential to change orbit. But resolution alone isn't enough. It's the combination of revisit capacity, high-resolution imagery, and analytics that enables us to support the Space Force mission to safeguard U.S. assets and maintain persistent awareness of what's happening in orbit. This is one of several Space Domain Awareness contracts we've won recently, from filling collision avoidance blind spots for the Office of Space Commerce to enabling rapid threat response through Space Force's Apollo Accelerator. The message is clear: when it comes to seeing what's happening in orbit, Vantor delivers. 📸 While not specific to Vantor’s work with the U.S. Space Force, this visual showcases Vantor’s space-to-space imaging capabilities. The bottom-right shows the Vantor physics engine user interface, which is the software we use to assess imagery feasibilities and plan our collections. On the left is a non-Earth image of TRACERS-2, a spacecraft that is part of a NASA mission to study how the Sun's energy interacts with Earth's magnetic field. On the top-right you can see intelligence about the spacecraft’s orientation and velocity, which we deliver with every space-to-space image to help analysts understand how the object is behaving. 

79

5 Comments

“But what about NATS?” - I hear this all the time—especially after my recent posts on MQTT. 𝐅𝐢𝐫𝐬𝐭: No single message broker or protocol is a silver bullet. Neither MQTT nor Kafka nor AMQP nor NATS. Never! 𝐒𝐞𝐜𝐨𝐧𝐝—𝐚𝐧𝐝 𝐭𝐡𝐢𝐬 𝐢𝐬 𝐰𝐡𝐲 𝐈’𝐦 𝐰𝐫𝐢𝐭𝐢𝐧𝐠: Synadia, the original donor of NATS, has told the CNCF it plans to withdraw the project, re-license it under the Business Source License (BUSL), and reclaim community assets such as nats.io and the GitHub organization. Industry experts (e.g., Alex Kritikos) have warned me for years that NATS wasn’t truly open source / had open standards. Now I see their point. 🔗 Full CNCF statement: https://lnkd.in/eAwxEdBe

39

12 Comments