Heavy Press for Domestic Manufacturing

This week: vibe coding manufacturing apps (paid), heavy press program, novel lithography and EUV masks, autonomous control AI agents for acid gas removal, thermodynamic computing, plastic alternatives

Nov 02, 2025

∙ Paid

Shop Talk

Capturing this week’s zeitgeist

The strategic push for massive press capabilities, first established during the 1950s “Heavy Press Program,” was a US military imperative to build superior aircraft. The goal was to replace the slow, labor-intensive “bits and pieces” assembly method with large, single-piece forged components, resulting in airframes that were significantly stronger, lighter, and faster to produce. This 70-year-old principle has been revived in the civilian sector through gigacasting. Pioneered by automakers like Tesla and enabled by machine vendors like IDRA Group and LK Machinery, this modern method uses enormous high-pressure die-casting presses to manufacture entire vehicle chassis sections as a single part. This consolidation is a manufacturing simplification (remove any unnecessary steps!), eliminating hundreds of individual components and welds, slashing production costs and time, and creating lighter, stiffer bodies ideal for electric vehicles.

The remains of a massive 12,000 ton press now a rusting frame standing outdoors behind a chain-link fence on a bright blue day. — A glimpse into the massive scale of 20th-century steel production at the Rivers of Steel National Heritage Area. These images show artifacts from the former Homestead Works, including a plaque detailing the history of the 12,000-ton press that forged armor for WWII battleships, a large vessel used for transporting molten metal, and the imposing frame of a 12,000 ton heavy industrial press.

Recognizing this as a critical, next-generation industrial capability, nations are now investing heavily to build domestic capacity. Just as the heavy press was a matter of national defense in the 1950s, gigacasting is now a matter of national economic and manufacturing security. This is evidenced by the Government of India’s recent Request for Quotation to study the forging capability.

📅 Next up: 2025 SEMA Show, November 4-7, Las Vegas, NV, USA where RealTruck is collaborating with Mattel’s Matchbox brand to create a one-of-a-kind, heavily modified Jeep Wrangler to be unveiled during SEMA Live on Tuesday, Nov. 4 from 1-1:20 p.m. PST at RealTruck’s Booth #50033 and Toyota debuts all-electric “bZ Time Attack” AWD Concept.

Assembly Line

This week’s Industry 5.0 breakthroughs and frontier technologies of the built world.

🐟🦾 How Shinkei Is Reshoring America’s Seafood Supply Chain

See my previous commentary on Alaska fish processing

🗣️ Mastercam’s Nand Shivkumar on the Challenges and Benefits of AI /YouTube/ instilling a “cool factor” to attract new talent into the industry.
🗣️ From cutting-edge grippers to Gemini’s world-building vision – insights from Schunk’s CTO and Google DeepMind’s Phillip Lippe /The Industrial AI Podcast/

☢️ The Risky Movement to Make America Nuclear Again

A wave of Silicon Valley startups, led by Oklo, is promoting advanced small nuclear reactors, particularly sodium-cooled fast reactor designs, as a transformative technology to power AI data centers and industries. The readiness of this technology is virtually nonexistent; Oklo’s own design remains unlicensed after failing its initial NRC safety review in 2022, and experts warn the underlying fast-reactor technology is historically dangerous and prone to sodium fires, refuting claims of passive safety. Despite this lack of technical validation, the business impact has been explosive, with a dealmaking frenzy among tech giants and Oklo’s market value soaring to $20 billion.

💿 Substrate Is Seeking to Crack ASML’s Dominance

A secretive US startup, Substrate, has introduced a novel lithography approach using a particle accelerator to generate shorter-wavelength X-rays, a technology it claims is cheaper and smaller than competing systems. This directly challenges the current industry standard of Extreme Ultraviolet (EUV) lithography monopolized by ASML, and Substrate also aims to become a vertically-integrated American foundry, competing with manufacturers like TSMC. In terms of readiness, the company states it has a production-quality 300mm wafer tool capable of printing 12-nanometer features, claiming this resolution is equivalent to the 2nm node and comparable to the latest High NA EUV machines. The potential business impact is to break Rock’s Law (observation that the cost of building a leading-edge semiconductor fab has roughly doubled every four years, with facilities rising from over $5 billion in the early 2010s to around $25 billion today) by reducing leading-edge wafer costs by an order of magnitude (from a projected $100,000 to $10,000).

Read more at Bloomberg, Tom’s Hardware, and discuss on X

💿 FST’s subsidiary, Isol, has supplied its next-generation EUV mask inspection equipment, FREM, to Japan’s largest photomask manufacturer /The Elec/

Breaking the Bottleneck

Breaking the Bottleneck | Getting to Success with Manufacturing AI

Breaking the Bottleneck is a weekly manufacturing technology newsletter with perspectives, interviews, news, funding announcements, manufacturing market maps, 2025 predictions, and more…

11 days ago · 6 likes · 1 comment · Aditya Raghupathy

New Product Introduction

Highlighting new and innovative facilities, processes, products, and services

🏭🛢️🧠🔁 Aramco and Yokogawa Achieve a Major Milestone with Commissioning of Multiple Autonomous Control AI Agents at Major Gas Facility

Yokogawa Electric Corporation, a leader in process automation technology, announces it has achieved a historic milestone in the deployment of industrial artificial intelligence (AI) solutions with Aramco, one of the world’s leading integrated energy and chemicals companies. Multiple autonomous control AI agents have been successfully implemented by Yokogawa at Aramco’s Fadhili Gas Plant in the Kingdom of Saudi Arabia to enhance operational efficiency.

The AI solution developed by Yokogawa uses multiple, coordinated AI agents of the Factorial Kernel Dynamic Policy Programming (FKDPP) reinforcement learning-based AI algorithm to directly and autonomously control and optimize acid gas removal (AGR) operations at the Fadhili Gas Plant.

📏 Lumafield Introduces Auto-Dimensioning, GD&T Suite for Accurate, NIST-Traceable 3D Measurement

Lumafield, the leader in industrial X-ray CT technology, announced Auto-Dimensioning, a robust new capability that provides engineers with easy, automated methods for Geometric Dimensioning and Tolerancing (GD&T). With Auto-Dimensioning, Lumafield’s CT scanning technology now recognizes geometric features and measures geometric and dimensional tolerances such as flatness, parallelism, and profiles. The system automatically extracts internal and external dimensions from a single, non-destructive scan. Auto-Dimensioning builds on Lumafield’s new NIST-traceable calibration process, a workflow that uses a reference artifact with a documented calibration chain. As a result, measurement tasks that previously took hours with a CMM or OMM are reduced to a matter of seconds.

🦾🤖🦿 How China built a humanoid robot that’s cheaper than your phone

Chinese startup Noetix Robotics has developed an affordable humanoid robot named Bumi, priced at approximately $1,380, comparable to a high-end smartphone. This low cost, which has led to strong initial sales in China, was achieved by fundamentally rethinking design and production. The company pursued vertical integration by designing critical components like control boards in-house, utilized a lightweight structural redesign with composite materials and targeted metal reinforcement to reduce weight and subsequent motor/battery costs, and established its supply chain almost entirely within China to leverage manufacturing density and reduce logistical delays. Bumi is primarily positioned as a 3.1-foot-tall social companion and open-programming learning platform for education and entertainment, marking a shift towards making humanoid robotics accessible to mainstream consumers.

🆕 NEO The Home Robot Order Today /1X on X and YouTube/
🗣️ “It’s hard for the public to see real progress when most humanoids today are actually teleoperated behind the scenes without disclosure” /Brett Adcock on X/
🗣️ “If we’re not at parity on at least one, it’s not going to pencil out. Full stop. It’s just not economically viable. How could it be, when you can literally hire a person? Humans are SO absurdly good.” /Andrew McCalip on X/

♨️🖥️ Thermodynamic Computing From Zero to One

Extropic is developing a new thermodynamic computing architecture to address the significant energy constraints limiting the scaling of artificial intelligence. Their innovation, the Thermodynamic Sampling Unit (TSU), is a novel type of probabilistic computer that performs sampling from complex probability distributions rather than traditional processing. This hardware, which utilizes highly energy-efficient “pbits” (probabilistic bits) made from standard transistors, is designed to minimize energy waste from data communication. When combined with Extropic’s new generative AI algorithm, the Denoising Thermodynamic Model (DTM), the TSU is simulated to run AI workloads using orders of magnitude less energy than current GPU-based systems, aiming to remove energy as the primary barrier to AI progress.

Read more at Extropic and discuss on X

🚙🔋 Hyperdrives cracks the ‘holy grail’ of EV motor cooling with manufacturable direct-conductor tech /Tech.eu/
🧲 SpinDrive raises growth funding to scale magnetic levitation bearing technology for more efficient, cleaner industrial machinery /Cision/

How Tim Cook Evaded Disaster at Apple This Year /WSJ/ Foxconn Industrial’s Shares Hit Record High After AI-Driven Earnings Surge /YiCai/ while Nvidia CEO Jensen Huang, Samsung Electronics Chairman Jay Y. Lee and Hyundai Motor Group Executive Chair Euisun Chung make a toast at a fried chicken restaurant in Seoul /CNN/

Business Transactions

This week’s top funding events, acquisitions, and partnerships across industrial value chains.

🪙🤝 The Ryerson-Olympic merger: Implications for metal fabrication /The Fabricator/

🇦🇺⚗️ Seaweed startup Uluu attracts $16m to scale world-first plastic alternative

Uluu has raised AU$16 million in Series A funding to build a demonstration plant and scale its technology transforming seaweed into a natural alternative to plastic to industrial production. The round was led by German growth investor Burda Principal Investments with support from Main Sequence, Novel Investments (the family office of one of the world’s largest textile groups), Startmate, and a consortium of leading impact and family investors including Fairground and Trinity Ventures.

Uluu’s next-generation materials perform like conventional plastics and can be processed using existing plastic manufacturing equipment. Unlike plastic made from fossil fuels, they are reusable, recyclable, home compostable and marine biodegradable—breaking down naturally without releasing microplastics. They’re strong, lightweight, waterproof and non-toxic, while being climate positive at scale.

🇺🇸🚜 Agtonomy Secures $18 Million Series B to Accelerate Physical AI Adoption in Off-Road Industries

Agtonomy, a leading software and services company specializing in automation and physical AI solutions for agriculture and land management, closed an oversubscribed Series B funding round totaling $18 million based on its proven commercial traction and growth potential. This investment was led by DBL Partners, a pioneering double bottom line venture capital firm renowned for early investments in Tesla, SpaceX, and Farmers Business Network, with participation from new investor Nuveen, one of the world’s largest owners and operators of agricultural land—managing over 2 million acres globally. Existing investors Autotech, Allison Transmission, Rethink Food, and Black Forest Ventures also participated in the round.

Agtonomy will use the new capital to advance its physical AI platform, expand equipment integrations with OEM partners, and scale commercial deployments across agriculture and adjacent off-road industries. These investments will enhance AI-driven fleet intelligence and smarter workflows, helping operators cut labor costs and improve safety and sustainability in orchards, vineyards, fields, and managed green spaces.