Top all-time · #11800
zach's tech blog
Personal blog · zach.behardwarestartupsai
zach's tech blog is a personal blog at zach.be about hardware, startups and ai. Since 2024, 6 different Hacker News readers have linked it 19 times, most often “Why is it so hard for startups to compete with Cadence?”. It is still writing: the last post was on 2026-10-13, and you can follow it by feed.
Too few sampled comments (4) to say how often HN recommends it.?
Why HN links it
HN commenters link zach.be posts on AI and unconventional computing hardware, such as Groq's limits at frontier-model scale and Extropic's pivot away from superconductors. They cite them as sources for their own points.
Summarised by Claude from a sample of 4 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
building chips and writing words.
Most-linked posts?
- Why is it so hard for startups to compete with Cadence?3 people · 3 mentions · first linked 2025-07-01
- Tensordyne is making AI compute more efficient using logarithms2 people · 2 mentions · first linked 2026-06-17
What Hacker News says?
Totally saw this one coming! [1] I think one major challenge they'll face is that their architecture is incredibly fast at running the ~10-100B parameter open-source models, but starts hitting scaling issues with state-of-the-art models. They need 10k+ chips for a GPT-4-class model, but their optical interconnect only supports a few hundred chips. [1] https://www.zach.be/p/why-is-everybody-talking-about-groq
Comment linking /p/why-is-everybody-talking-about-groq
on “Groq CEO: 'We No Longer Sell Hardware'”Submitted YC is wrong about LLMs for chip design as a story · 322 points
HN discussionthere is also Normal Computing[0] that are trying different approaches to chips like that. Anyway these are very difficult problems and Extropic already abandoned some of their initial claims about superconductors to pivot to more classical CMOS circuits[1] [0]: https://www.normalcomputing.com [1]: https://www.zach.be/p/making-unconventional-computing-practi...
Comment linking /p/making-unconventional-computing-practical
on “Extropic is building thermodynamic computing hardware”Submitted Why is it so hard for startups to compete with Cadence? as a story · 20 points
HN discussionThe big challenge when it comes to using FPGAs for deep learning is pretty simple: all of that reprogrammability comes at a performance cost. If you're doing something highly specific that conventional GPUs are bad at, like genomics research [1] or high-frequency trading [2], the performance tradeoff is worth it. But for deep learning, GPUs and AI ASICs are highly optimized for most of these computations, and an FPG…
Comment linking /p/how-to-build-a-commercial-open-source
on “FPGA Architecture for Deep Learning: Survey and Future Directions”Submitted Tensordyne is making AI compute more efficient using logarithms as a story · 6 points
HN discussionI'm curious what the motivation is here -- unfortunately, the dev blog is all in Chinese and I can't read it. If it's mostly to show a proof-of-concept of LLMs on a FPGA, that's awesome! But if this is targeting real-world applications, I'd have concerns about price-to-performance. High-level synthesis tools often result in fairly poor performance compared to writing Verilog or SystemVerilog. Also, AI-focused SoCs l…
Comment linking /p/how-to-build-a-commercial-open-source
on “Swan – A Lightweight Language Model Execution Environment Using FPGA”Submitted Why is it taking so long to build new IP cores? as a story · 4 points
HN discussionSubmitted How the hell is Groq raising more money? as a story · 3 points
HN discussionSubmitted Why is it so hard for startups to compete with Cadence? as a story · 3 points
HN discussionSubmitted Why is everybody talking about Groq? as a story · 3 points
HN discussionSubmitted Tensordyne is making AI compute more efficient using logarithms as a story · 2 points
HN discussionSubmitted Crossbar's new security chip isn't secure as a story · 2 points
HN discussionSubmitted Why is it so hard for startups to compete with Cadence? as a story · 2 points
HN discussionSubmitted Why hasn't there been a big winner in edge AI chips? as a story · 2 points
HN discussionSubmitted Was I wrong about Etched? as a story · 1 points
HN discussionSubmitted Why did Untether AI fail? as a story · 1 points
HN discussionSubmitted What's the difference between all these quantum computers? as a story · 1 points
HN discussionSubmitted What's the difference between Extropic, Normal Computing, and D-Wave? as a story · 1 points
HN discussion
