Jinfeng Huang

I work closely with several leading autonomy and robotics companies in the Bay Area 🚗🤖 and regularly connect with engineers across planning, perception, simulation, and robotics systems. One trend I’m seeing across many teams: strong systems engineers are increasingly in demand. The stack (often C++ / Python) matters, but what matters more is deep domain experience in robotics or autonomous systems, particularly around safety, evaluation, validation, and fault handling. If you’re working in this space or curious about what’s happening across AV and robotics, feel free to reach out. #robotics #autonomoussystems #physicalAI

28

4 Comments

So proud to share what our team has been building! Mark shared our exciting announcements at Meta Connect. As we're focused on building the deepest lineup of AI glasses in the world, here’s what we announced: 😎 Ray-Ban Meta (Gen 2): Up to 8 hours of battery life, 3K Ultra HD video capture, and new styles. 🚴 Oakley Meta Vanguard: The Next Era for Performance AI glasses. Built for athletes who want to capture every moment, push their limits, and stay connected to their goals. 👓 Meta Ray-Ban Display: The most advanced AI glasses ever built, featuring a visual display and controlled by the revolutionary Meta Neural Band. #MetaConnect #AI #AIglasses #innovation

156

4 Comments

The end of human drivers? Tried Tesla Robotaxi in the Bay Area for the first time, 8.8 miles from Meta SF to Daly City station for $18. Unlike Waymo, Tesla’s Robotaxi can take the freeway. The ride was smooth, and the decision-making felt far closer to human than “bot-like.” What surprised me most: it’s just a regular Model Y. With the built-in rear screen, any Tesla could potentially be activated as a Robotaxi. That’s scalability at an entirely different level. Waymo might have an early lead, but Tesla’s vision-only approach gives it massive coverage and faster rollout potential. If this scales, Uber, Lyft, and even car ownership itself could be disrupted faster than expected. This is exactly why I quit my job to go all-in on ElasticDash We’re entering a new industrial revolution, the real question is: will you shape it, or be forced to adapt?

15

2 Comments

The Code Review reported on NVIDIA and Jensen Huang's announcement of the Drive AGX/Alpamayo autonomous technology, and what it means for industry players. With a special Fetii mention! "The announcement arrives at a moment when mobility players like Uber, Lyft, Waymo, and newer models such as high-capacity group transportation platform Fetii are defining increasingly distinct approaches to how people move." As NVIDIA and others standardize and open up autonomous vehicle technology, the cost curve will continue to compress and adoption will accelerate across OEMs. Autonomy won’t remain a novelty or a premium feature, it will become table stakes. The real advantage will belong to platforms that can turn that technology into sustained, high-utilization demand. That’s where Fetii is uniquely positioned. By focusing on high-capacity, repeatable group movement, Fetii and its partners are built to extract real economic value from autonomous fleets as they scale.

23

3 Comments

💨 How fast can an autonomous vehicle think? With Alpamayo 1, NVIDIA's 10B-parameter chain-of-thought reasoning model, the distilled version can reason in real time. Hear Marco Pavone, Yan Wang, Yurong You, and Wenhao Ding from our AV Research team break down Alpamayo 1 and what's next for reasoning in autonomous driving. 🔁 Watch the replay: https://bit.ly/4qDKysS

7

Why I'm Flying to Work with my Remote Engineers (Instead of Finding a Bay Area Cofounder) Founder Update Three months ago, I started searching for a cofounder for my robotics startup in California. The ideal candidate profile was clear: 10x engineer, top-tier school credentials (helpful for fundraising pre-traction), located in the SF Bay Area. After three months of searching—including creative outreach that generated half a dozen face-to-face interviews—I found exactly zero cofounders. The Expiration Date Here's the thing about cofounder searches: they have an expiration date. At some point, the business needs to move to its next phase, with or without the perfect cofounder. I hit that point two weeks ago during a manufacturing trip to China. While there, I did some math. The cost of flying to China and staying for two weeks was two orders of magnitude less than paying a single month's SF Bay Area salary for an engineer ($2K/month vs $160/year + $2K/month benefits + $25K capital rais bonus + equity). That's when I realized I was optimizing for the wrong constraint. Splitting the "Ideal" Cofounder I was looking for someone who was both a 10x engineer and had top-tier credentials. But what if I split that persona into two separate hires? New approach: Hire 1: Remote high-performer via Upwork - $20-80/hour (2x to 4x cheaper than SF startup salaries) - Autonomous, enthusiastic, hard worker - Location: anywhere Hire 2: Part-time advisor with tier-1 credentials - Equity only - Primary role: help with fundraising credibility - Location: anywhere My wife is okay with me traveling, so I dropped the location filter entirely. The Problem With Remote at This Stage Here's where it gets tricky. I'm still at the very early MVP stage. The ideation and trial-and-error loop is brutally fast—especially because I'm using Claude Code to implement ideas. (I use Claude for everything in MVP prototyping, and it's been fantastic.) The problem: that fast iteration loop doesn't work well with remote. If I send specs in the evening, get implementation in the morning, test during the day, and repeat—we're already adding 12+ hour delays to each cycle. When you're running multiple cycles per day in-person, going remote feels like moving through molasses. The Solution: Hybrid "Out-Person" + Remote Here's what I'm planning: - Fly to the freelancer's city - Stay in a hotel for 1-2 weeks - Bring hardware and MVP prototype with me - Work face-to-face to get them to 100% productivity - Return home - Work remotely until the next in-person sprint is needed - Repeat The cost structure makes sense. A week or two of travel + accommodation + focused work is still 10x dramatically cheaper than SF salaries. And I can work from anywhere with my laptop. What do you think about it? - Has anyone else tried this hybrid approach? How did it work out? - What are the failure modes I'm not seeing? - Is this sustainable long-term?

26

11 Comments

https://lnkd.in/gJHUXv49 If you want to truly understand how modern LLMs work under the hood, Stanford’s CS336: Language Modeling from Scratch is one of the best resources out there. Instead of treating LLMs as black boxes, the course walks through the full stack: • tokenization & dataset pipelines • transformer architecture implementation • attention & positional encoding • training dynamics & scaling laws • distributed training and GPU efficiency • alignment and post-training You end up implementing a full language model training pipeline yourself. It’s rare to see a course that connects model architecture, training theory, and systems optimization this cleanly. Highly recommended for anyone interested in LLM research, systems, or building foundation models.

2

Heading to DesignCon in Santa Clara this week (Feb 24-26)! I'm building OptixLog: a knowledge graph platform that connects design, manufacturing, and test data for semiconductor teams. We're solving the data fragmentation problem that makes it hard to trace decisions from simulation to production. If you're attending DesignCon or in the Santa Clara area this week, I'd love to grab coffee and chat about: Knowledge graphs for hardware workflows How LLMs can actually reason across your full process (not just isolated tool outputs) Where data traceability breaks in semiconductor R&D Or just swap notes on building in the hardware space Drop me a message here or email me at founder@optixlog.com. Always down to learn from folks shipping silicon. See you there!

30

1 Comment