Advertise on podcast: electronics.dev

Join Seve (founder of tscircuit) and Matt (founder of atopile) as they venture deep into the rapidly changing world of electronics, where hardware is finally catching up to software. This episode tackles the growing movement of designing circuit boards using code, not schematics, and explores how compiler-style toolchains are revolutionizing firmware, FPGA workflows, and hardware reliability. They also explore: * Wi-Fi HaLow: A deep dive into the 900MHz standard that offers kilometer-long range for drones and IoT, and how it challenges LoRa. * The “Monks” of SQLite: Why the SQLite team is considered the “monastery” of open source and their commitment to no API breaks until 2050. * Semver vs. Reality: Why Semantic Versioning breaks down in TypeScript environments and Seve’s “Pragmatic Versioning” alternative. * The Future of Connectivity: How Starlink’s direct-to-cell technology is effectively placing cell towers in space. * AI & DSLs: Whether LLMs will end the era of new programming languages in favor of Domain Specific Languages or TypeScript supremacy. Whether you are a firmware engineer looking at Rust, a hardware designer interested in long-range wireless, or a developer managing complex dependencies, this discussion offers a fresh perspective on the intersection of code and hardware. 00:00 Hiking Yosemite and the Adirondacks 06:28 Rust in firmware 07:06 Truth Discovery 10:21 Safety in Embedded: Rust vs. C/C++ 12:42 Airbags & Real-Time Constraints 14:55 Haskell, OCaml, and Functional Programming 16:44 Why Semver doesn’t work for TypeScript 17:51 Ecosystem development: 200 Releases a Day 26:42 Using OCaml for FPGA Design (Jane Street) 29:29 The “Monks” of SQLite & 2050 Compatibility 33:40 Will AI stop the invention of new languages? 39:05 Wi-Fi HaLow: 900MHz Long-Range Wi-Fi 43:50 Wi-Fi HaLow for Drones & Antenna Physics 48:07 Starlink Direct-to-Cell Technology This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Seve (founder of tscircuit) and Matt (founder of atopile) unpack one of the most radical ideas in modern electronics: reversible computing. As Moore’s Law slows down, the biggest bottleneck is heat — but what if computation itself could be rethought to conserve energy instead of wasting it? In this episode, they explore: * The principles of reversible/thermodynamic computing * How MOSFET charge recycling offers a real-world analogy * Why reinventing chip architectures (CUDA, ISAs, RISC-V) is so challenging * The promise and pitfalls of open-source hardware ecosystems * Lessons from Meta Connect’s failed demos and the future of hardware/software integration * Why “hard tech” founders still matter in a software-driven world For hardware engineers, embedded developers, and anyone curious about the future of chips, this episode connects deep insights into electronics with the big questions of computing’s next era. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Apple’s new iPhone Air might look like a futuristic, ultra-thin smartphone, but under the hood, it’s closer to an Apple Watch strapped to a giant battery. In this episode, Seve (founder of tscircuit) and Matt (founder of atopile) break down the wild engineering behind making phones slimmer than ever: iPhone Air Deep Dive: * Why the iPhone Air’s PCB layout looks more like a smartwatch than a phone * How Apple carved down the circuit board to a tiny “plateau” under the camera * The real limits of *batteries, casings, and glass thickness* * Why reducing *just 1 mm* of thickness is brutally hard * Tricks like *embedded resistors* inside the PCB stack-up * Why thinner phones aren’t just aesthetic — they change the entire board design Beyond the iPhone — Startup Life & Silicon Valley: * What “honest tells” are — from biology to founder culture * The strange power moves of Silicon Valley investors and founders * Why outfits are status signals in tech * Lessons from Reddit’s messy founder history * How investor–founder dynamics shifted in the past 15 years * The psychology of saying “no” to meetings (and why it makes you work harder) If you’ve ever wondered how Apple keeps packing a supercomputer into a device smaller than your wallet or what it’s really like building hardware startups in Silicon Valley, this episode blends both worlds. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Auto-routing sounds simple: connect the traces, hit “go,” and let the software handle the rest. But in reality, PCB auto-layout is one of the hardest problems in electronics. In this episode of electronics.dev, Seve (founder of tscircuit) and Matt (founder of atopile) break down why auto-routing is so difficult: from impedance matching and creepage rules to high-speed signal integrity and noise coupling. They also explore the new wave of AI-driven solutions and whether they can finally deliver on the promise of hands-free PCB layout. Along the way, the duo dives into: * Why traditional auto-routers often fail on complex boards * The trade-offs between speed and quality in modern routing algorithms * How engineers use SAT solvers and constraint systems to optimize pin assignments * What tools like Quilter and DeepPCB are getting right (and wrong) * Why AI-assisted optimization might work better than full AI-generated layouts * How the future of PCB design could be shaped by deterministic solvers + AI refinement If you’ve ever wondered why PCB design still feels like an art form, and whether AI will finally make auto-routing reliable, this conversation offers an inside look at the problem and the potential solutions. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Custom chip design is no longer reserved for billion-dollar tech giants. In this episode, Seve (founder of tscircuit) and Matt (founder of atopile) explore how advances in tooling, prototyping, and the EDA industry are making custom silicon accessible to startups, makers, and engineers like never before. You’ll discover: * How Tesla’s early battery management challenges led to unique custom chip solutions * The surprising economics of a $7K run for 5,000 prototype chips * Why most microcontrollers still use massive 90nm processes, and why that works * Analog vs. digital scaling: where Moore’s Law still applies and where it doesn’t * Funding trends in hardware startups, from Diode to SnapMagic * The state of open-source electronics after the OpenSauce conference * CNC milling “micron accuracy” claims: myth vs. reality * Why PCB milling often isn’t worth the time for serious prototyping Whether you’re a hardware engineer, startup founder, or electronics enthusiast, this conversation pulls back the curtain on the tools, economics, and decisions shaping the next generation of hardware. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

AI Isn’t Coming For Electronics Design. It’s Already Here. Matt (founder of Atopile) and Seve (founder of TSCircuit) return with a deep dive into the rapidly shifting world of electronics and mechanical design. In this episode, they discuss how AI is reshaping every part of the hardware stack: from PCB layouts to robotic assembly. What you'll learn today: * Why "arms in a box" robotics startups are booming * How Meta’s world models might replace control loops * The explosive growth of synthetic data for training AIs * Why OpenCascade and legacy CAD tools are holding engineers back * The rise of custom, code-native CAD kernels like Zoo * The power of constraint-based design thinking * Could text-to-CAD finally be useful in real workflows? * Why betting on “X as code” years ago is paying off now Whether you’re a hardware engineer, an AI researcher, or a curious builder, this conversation reveals the core changes happening right now in how electronics are designed. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

In this episode of electronics.dev, Seve (founder of tscircuit) and Matt (founder of atopile) break down a quiet revolution in electronics: designing circuit boards with code and compiler-like workflows, not traditional schematics. They dive into: * Why traditional EDA tools limit innovation * How constraints, templates, and AI are unlocking more expressive, testable circuit design * The surprising overlap between web layout (CSS) and PCB layout * How tools like atopile let you think in relationships, not in traces * What “spatial reasoning” really means for engineers, and why half of us may be bad at it * Why autonomous layout and reference design matching will change everything Along the way, the duo shares reflections on: * The enduring power of sketching in engineering * Using AI as a design partner (and critique engine) * How mechanical and electronics design differ, and what each can learn from the other * Why circuit fabrication is standardized, but placement logic is still tribal knowledge If you’ve ever written hardware configuration code, fought with schematic tools, or dreamt of a world where hardware is as programmable and composable as software, this episode is for you. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Seve (founder at tscircuit) and Matt (founder at atopile) tear into the wild intersection of geopolitics and hardware as they explore a North Korean KN-23 ballistic missile teardown. This episode reveals how Western components still end up in restricted military tech, and how second-hand supply chains blur the line between hobby electronics and weapons of war. But the conversation doesn’t stop there. From chip smuggling to the future of prototyping, this episode explores how electronics shape everything from missiles to laundry robots. What You’ll Learn Today: * How NXP, Analog Devices, and Fairchild chips wound up in a North Korean missile * The surprising link between PlayStation 2 and military supercomputers * Why chip reverse engineering is rarely worth it, unless you're a rogue state * How China is trying (and struggling) to close the chip tech gap * Why ASML is a single point of failure for the global chip industry * The future of humanoid robots and open-source robotics startups * A new “jumperless breadboard” that redefines prototyping * A wild idea for a fully automated PCB assembly shop Whether you're into defence tech, microcontroller ecosystems, or the nerdy tools reshaping prototyping and manufacturing, this episode will blow your mind! This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Are we finally close to fabbing real PCBs at home? In this episode, Seve (tscircuit) and Matt (atopile) go deep into the state of DIY PCB fabrication: why it feels like the early days of 3D printing, what's still holding it back, and what could push it over the edge. We talk lasers, vias, copper, tariffs, and some surprising hacks people use to prototype faster without sending things to China. What You’ll Learn Today: * How fiber lasers are being used to blast copper off PCBs * Why vias are still the bottleneck for DIY PCB prototyping * What a $200K laser ablation machine actually does * How one engineer claims to make PCBs with no chemicals at all * Whether DIY sintered copper vias might actually work * How U.S. tariffs sparked a wave of fake “local” PCB fabs * Why 3D printing and PCB fab might follow the same path * What still needs to happen before in-house PCB manufacturing is truly usable Whether you're building hardware, designing boards, or just wondering what’s next for fast prototyping, this is one of our most technical and fun episodes yet. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Join Seve (founder at tscircuit) and Matt (founder at atopile) for a no-fluff conversation about the future of electronics design. In this episode, the founders dive deep into: * Why designing PCBs with code is better than legacy schematic tools * How Chinese factories like JLCPCB dominate with software automation * The hidden costs of tariffs — and why they don’t actually help US manufacturing * Why open-source hardware and standards like ODB++ and IPC are a mess * Real talk on US vs. Chinese PCB fabs, and what needs to change Whether you're a hardware engineer, startup founder, or just curious about how electronics are made, this episode is packed with firsthand insights, rants, and laughs. Bonus topics: * Automating your business taxes with AI * What “open source” really means in 2025 * The surprisingly painful experience of launching a landing page This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Seve's original post Part 1 episode This episode continues where the last one left off, diving into Part 2 of Seve's blog post “13 Things I Wish I Knew Before Building an Autorouter.” 🧠⚡ Seve (founder of tscircuit) and Matt (founder of atopile) explore the deep technical challenges of auto-routing printed circuit boards using code, compiler toolchains, and caching, rather than traditional schematics-based tools. In this second half, they unpack: * Why caching is crucial for speeding up PCB auto-routing * Visualization as a debugging superpower * The power of PyInstrument and visual profiling tools * Why recursive functions and Monte Carlo methods often fail in optimization * Game dev tricks (like greedy A*) adapted for PCB pathfinding * The difference between grid-based vs intersection-based routing math * Why keeping results grounded in physical space (like millimeters) makes debugging easier * Using animation to catch stupid behavior before it goes live * How a meta-router manages multiple strategies in parallel * Insights from CNC machines and high-fidelity spatial modeling This episode is packed with Seve and Matt's hands-on experiences, hard-won insights, and sharp advice for anyone building or using modern, code-first electronics design tools. 🛠 Whether you're a PCB engineer, systems architect, or startup founder in the hardware space, this conversation is pure gold. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

In this special guest episode, Matt (founder of Atopile) talks with Tyler Maran, founder and CEO of OmniAI, about how traditional OCR gives way to AI-native tools that actually work. OmniAI uses cutting-edge vision-language models to parse datasheets, extract data from messy PDFs, and interpret charts, with accuracy levels that leave legacy OCR in the dust. This episode dives deep into how modern AI models can read and understand electronics documentation, turning datasheets into structured, machine-usable formats like Markdown, HTML, and JSON. Forget brittle extraction rules and regex hacks — this is about intelligent parsing. 💡 What you’ll learn today: * How OmniAI uses vision-language models to parse 600 million datasheet pages * Why LLMs are outperforming traditional OCR in accuracy and cost * The insane complexities of converting PDF charts into usable circuit data * How AI models handle errata, tribal logic, and embedded graphs * The concept of "thinking models" vs "simple models" for different document types * Why designing PCBs with code gives you a feedback loop, just like software * What “agent mode” looks like for parsing hardware documentation * Why human-level AI might be running on iPads in 2 years * How Matt uses Atopile and Cursor to compile hardware like software Plus: patent law hacks, robotic bartenders, anarchist hackerspaces, and how laziness can be a superpower in engineering. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Join Seve (founder of tscircuit) and Matt (founder of atopile) as they analyse OpenAI's impressive GPT-4.1 release and contrast it with Meta's controversial Llama 4 launch. In this episode, our hardware and AI experts explore: * OpenAI's brilliant strategy of releasing three GPT-4.1 variants: standard, mini, and nano * How the tiered approach allows developers to choose the right model for their specific needs * Why GPT-4.1's instruction-following capabilities make it ideal for agent-based applications * The stark contrast with Meta's Llama 4 release, which has failed to live up to its benchmark claims * Why no one has been able to reproduce Llama 4's claimed performance levels * The critical importance of single-GPU inference for edge computing applications The duo also discusses fascinating related topics: * The economics of AI deployment and why local compute makes sense * How tariffs are reshaping tech manufacturing decisions * The revolutionary potential of edge AI for robotics * Why latency requirements make on-device AI essential for advanced applications * The future of modular electronics for AI development Whether you're an AI developer, hardware engineer, or tech enthusiast, this episode offers crucial insights into the evolving landscape of AI models and their deployment strategies. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Subtitle: Two hardware founders explore how AI models are transforming electronics development, sleep science, and creative imagery Description: Join Seve (founder of tscircuit) and Matt (founder of atopile) as they dive into the AI revolution reshaping hardware development and everyday life. This dynamic conversation explores how cutting-edge AI models transform electronics design, personal health tracking, and creative expression. The duo unpacks: * How Gemini 2.5 Pro's massive 1 million token context window is revolutionizing hardware debugging and algorithm development * Why PCB auto-routing is finally becoming viable after decades of skepticism from hardware engineers * The viral phenomenon of "ghiblifying" photos with GPT-4o's image generation capabilities * Andre Karpathy's comprehensive sleep tracker experiment revealing which devices work (and which are "random number generators") * Why spatial hash indexing outperforms traditional quadtree structures for PCB routing algorithms * The shifting power dynamic between software and hardware development in the AI era Plus fascinating tangents into: * The technical marvel inside Oura Ring's tiny form factor * Creative ways to visualize technical concepts using AI image generation * How to use Repo-mix to dump entire codebases into AI systems for debugging * The untapped potential of AI-assisted hardware design workflows * Morning routines of engineers and viral "5 am routine" videos Whether you're a hardware engineer looking to leverage AI in your workflow, a sleep optimization enthusiast, or just curious about the bleeding edge of technology, this conversation offers unique insights into how artificial intelligence transforms our tools and creative possibilities. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev

Join Seve and Matt as they dive deep into the incredible story of Lilium, the billion-dollar electric vertical takeoff and landing (VTOL) startup that promised to revolutionize air travel—and then spectacularly crashed. In this eye-opening episode, you'll discover: * The shocking moment a multi-million dollar prototype caught fire * How startup culture clashed with aerospace engineering * The critical importance of rapid iteration in high-stakes technology * Why being caught between "startup" and "aerospace company" can be fatal * Insider stories from ground zero of an ambitious technological moonshot Matt shares firsthand experiences from his time as a mechanical engineer at Lilium, revealing: * The challenges of designing ultra-lightweight aircraft components * The complex world of aerospace prototyping * How a single patent can emerge from near-disaster * The psychological challenges of working in a high-pressure startup environment Whether you're an engineer, entrepreneur, or tech enthusiast, this episode offers a gripping narrative of innovation, ambition, and the razor-thin line between breakthrough and failure. Warning: This podcast сontains explosive (literally!) engineering stories and startup drama that you won't find anywhere else. Timestamps: 00:00 Claude 3.7 Sonnet 13:50 Lilium Story This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit electronics.dev