Max Hodak: Average Is Not Good Enough

Y Combinator · 2026-08-07 · 57 min
https://www.youtube.com/watch?v=Xc4klGbq8v8Video summary
Science CEO Max Hodak says Helix speeds execution, while a patient has read a 300-page novel with its retinal implant.
Science CEO Max Hodak introduced the company’s retinal implant, which uses infrared light and an implant beneath the retina to restore vision; one patient read a 300-page novel with it. At Startup School 2026, he argued that startup speed depends on unglamorous infrastructure: procurement, budgets, experiment-cost tracking, and internal software such as Science’s Helix. One wafer iteration cost $40,000, making clear why research spending must be measured. He described a four-stage hiring process—companywide votes, phone screens, homework, and interviews—and “IGEN reviews,” which use recurring employee feedback to assess performance. Hodak’s central message was that iteration speed shapes success, but founders cannot delegate judgment or rely on universal startup rules. In the Q&A, he advised raising twice the cost of the next key experiment, discussed AI’s uses in coding and regulatory work, and described the power, heat, and packaging challenges facing neural implants.
Chapters
- 0:00Infrastructure at Startups: Max Hodak’s Duke Brain-Computer Interface Research
- 1:02Science’s Retinal Implant and the Hidden Infrastructure of Deep-Tech Startups
- 3:25How a 17th Employee Buys Equipment—and How Infrastructure Builds Speed
- 7:03Experiment Costs in Helix and Hiring from a Startup’s Network
- 10:18Science’s Hiring Process: Companywide Votes and a 17% Phone-Screen Rate
- 13:01Science’s Four-Stage Hiring Process Tests Judgment, Horsepower, and Agency
- 16:06IGEN Reviews: Helix’s Four-to-Six-Week Feedback and the Case for Faster Iteration
- 18:53Weekly Learning, Operational Infrastructure, and the Founder’s Undelegable Judgment
- 20:43Why Startup Founders Cannot Delegate Judgment—and Must Act to Learn
- 22:47Action Produces Information: Company Systems as the Operating System
- 25:10Q&A: Formula SAE wins and clear thinking guide Science’s hiring
- 29:36Q&A: Neuralink judgment lessons, consciousness research, and opsins
- 35:11Q&A: Helix AI agents and custom hiring software
- 39:31Q&A: BCIs as longevity technology and safeguards for review votes
- 42:58Q&A: Fund the next experiment, raise twice its cost, and pursue revenue
- 47:46Q&A: Wireless retinal implants, packaging, and four stimulation approaches
- 52:15Q&A: Science’s 30/70 culture, six-country trials, and Helix’s contrarian purchasing system
What the full analysis includes
17 chapters · 88 key points · 17 section summaries · mind map
Preview of the first chapter · Infrastructure at Startups: Max Hodak’s Duke Brain-Computer Interface Research
The remaining 85 key points, section summaries and mind map are in the full analysis. Analyze this video to see everything →
This video is 58 min. The Free plan analyzes videos up to 30 min; plans from Plus cover videos up to 2 hours.
This is a Tier 1 public summary
Whether the chapter key points, section summaries and mind map are public is up to the person who shared it. Want the full analysis?Submit one yourself.
More from this channel
What Happens When Everyone Can Build SoftwareY CombinatorSupabase CEO Paul Copplestone says AI builders could drive 1.2 billion personal databases monthly, as Supabase surpasses 13 million users.
The Case For Data Centers In SpaceY CombinatorStarcloud CEO Philip Johnston says orbital data centers could add 20 GW of compute, with 88,000 satellites planned.
How To Design In The Agent EraY CombinatorPaper founder Stephen Haney says HTML/CSS-based design and human judgment can help teams ship faster without AI-generated sameness.
Patrick Collison: Is AI Breaking the Lean Startup Playbook?Y CombinatorStripe CEO Patrick Collison says new businesses on Stripe are up nearly 2x year over year, making this a remarkable time to start.
Why Robotics Still Isn't Solved - But Could Be Soon | YC Paper ClubY CombinatorYC Robotics Club researchers tackle robot memory, action reasoning, and dexterous tool use as startups seek practical deployment.
Related analyses
Why AI Could Turn Small Law Firms Into Billion-Dollar Businessesa16zEve CEO Jay Madheswaran says AI helped an LA law firm grow from 100 to 500 clients per attorney by automating plaintiff-law work.
OpenAI Dev Day 中文翻譯:Dots、GPT-6.1 Sol、Ultrafast、雲端 Codex 一次看懂電腦王阿達OpenAI Dev Day 推出 Dots、雲端 Codex 與 GPT-6.1 Sol,Ultrafast 每秒可生成 300 tokens。
E424: 用 AI 把自主權還給全盲的母親!當稀缺資源變成「算力競賽」,公平該怎麼重寫?|Peggy【塞掐 Side Chat】E424INSIDE 硬塞的網路趨勢觀察Peggy 用 AI 助理讓全盲母親五個月發出逾 800 則指令,也指出公部門導入 AI 最難的是法規與資料分級。
E428: 十分鐘就能開店做生意!微型賣家靠 AI「大逆轉」?|Pinkoi 共同創辦人暨執行長 Peter【塞掐 Side Chat】E428INSIDE 硬塞的網路趨勢觀察Pinkoi 執行長 Peter 推出 PinZap,讓微型賣家十分鐘開店,AI 也把競爭焦點推向產品力與品牌力。
2026/10/27(二)美股止跌了?加碼or觀望 Anthropic上市倒數 期中選舉恐衝擊AI?【早晨財經速解讀】游庭皓的財經皓角Anthropic 半年營收暴增 13 倍,AI 資料中心卻受融資、電力與美國期中選舉風險牽制。