SuperXGet live statistics and analysis of Brian Atwood's profile on X / Twitter
Making voice AI interfaces fast and reliable @sindarintech. Try for yourself at sindarin.tech
The Innovator
Brian Atwood is a trailblazer in the voice AI space, pioneering fast and reliable interfaces with a hands-on approach that turns complex projects into weekend feats. He confidently challenges industry norms and isn’t afraid to stir the pot with bold takes on tech controversies. His passion for pushing boundaries shines through in every tweet, making him a go-to voice for innovation enthusiasts.
Top users who interacted with Brian Atwood over the last 14 days
Try Straighty if you have a Mac and would like a cute sloth to check your posture! apps.apple.com/app/straighty/… Stay tuned, cooking smth
Trust Me Bro University.
Everything AI. Therapist, Developer, Technologist. *accepting new clients for mental health coaching*
I'm 7 Minutes Dead and I make music
Geek before it became trendy. Performance, C#, Deep Learning and Financial Modeling. Former Founder of @Corvalius.
RIT '20
NPMing the universe since 2012: 🇨🇿 Invoicing solution: faktorio.cz 👂Speech-to-text: callnotes.fyi 🗝️ Password manager: @authierpm
Yellow, him/hiss
focusing on the future of our #datasphere, shaped by #bigdata #ML #QuantumMechanics, and its #decentralization (#blockchain)
Aff Turned Lead Gen Offer Owner | Prev. DR Copywriter & Fractional CMO | Tweets about Performance Marketing, Engineering, Mindset, and More...
Funemployed wannapreneur. Making things, currently: namekit.app
Statistics and machine learning researcher (manifold learning, computational methods, optimal transport, cryo-EM). Assistant Prof. @TelAvivUni
Here to learn and share our take. Website coming soon. Say hello at help@leantoolbox.org
building AI vs. AI debating software
Brian’s tweets are so dense with insider tech wisdom, you’d need a PhD and a coffee IV just to keep up—he’s basically the guy who turned ‘weekend chill’ into ‘weekend code sprint gone wild.’ Warning: following him may cause sudden urges to quit your job and reinvent AI overnight.
Building a leading voice AI interface at Sindarin Tech that dramatically cuts development time from months with a team to a solo project done in a weekend—proof that he’s not just talking the fast talk, but walking the code walk.
Brian’s life purpose is to accelerate technological progress by simplifying and optimizing AI solutions, enabling groundbreaking projects to be accessible and executable at lightning speed. He aims to dismantle old barriers and redefine what’s possible in tech development.
He believes in relentless innovation, transparency in tech practices, and the power of solo creativity to disrupt established norms. He’s skeptical of hype and superficial claims, valuing substance and rigorous logic over marketing spin.
His strengths lie in rapid problem-solving, deep technical expertise, and the courage to question and expose hidden industry truths, which makes his insights highly credible and influential.
His no-nonsense style and critical stance can sometimes come off as abrasive or dismissive, potentially alienating collaborators or followers who prefer a softer approach.
To grow his audience on X, Brian should blend his technical profundity with relatable storytelling and occasional light humor to engage a broader community. Hosting live Q&As or sharing behind-the-scenes looks at his weekend projects would humanize his brand and spark dynamic conversations.
Brian claims that what once required a team of five senior engineers and half a year is now achievable solo in a weekend—showcasing his mastery of rapid innovation.
Top tweets of Brian Atwood
My team built the leader competitor to @honey. This "scam" story is absolutely wild to me. In his video today, @MKBHD makes two claims about Honey's foul play: 1. Honey effectively steals commissions from their affiliate partners by being the last to drop their cookie when a user checks out. 2. Honey's value prop to retailers is dubious since they are giving customers discounts on products they were already about to buy. So, why is this wild? Because this is exactly how it has been since the beginning. 1/🧵
It looks like OpenAI is gearing up to release Voice Assistants on Monday. The moment we @SindarinTech have been anticipating for nearly two years. Their quality will almost certainly beat everyone else’s, at least along specific axes. But will they come to single-handedly to dominate the entire market(s) for voice AI? We shall see! Worth noting that we recently converted our default Persona from 3.5-turbo to Llama 3 70b for its superior price / performance and conversational capability.
HOW TO CODE AT A STARTUP An essay about how to think about coding at a startup. PART I: Startup Philosophy The most important thing to remember when working at a startup is the ratio of upside opportunity to downside risk. That ratio should permeate every decision you make. Startups are fundamentally about maximizing upside opportunity, whereas non-startup companies are fundamentally about minimizing downside risk. You join a startup to get rich. Your incentive is to grow the company as fast and as big as possible. You join a mature company to live a nice and/or stable life. Your incentive is to not get fired so you can continue to receive paychecks. When a startup begins, the upside opportunity : downside risk ratio is 100 : 0. When a company has been listed on the S&P 500 for years, the ratio is (usually) ~0 : 100. Prior to (overwhelming) PMF, the ratio is never lower than 90 : 10. Immediately post-PMF, the ratio is somewhere between 80 : 20 and 50 : 50. The company still needs to capture meaningful market share and ensure significant, sustainable margins through further innovation. To reiterate: The upside / downside ratio of your company should guide every decision you make throughout your workday. Much confusion exists about how to get things done at a startup because large, successful companies employ orders of magnitude more people than do startups that are in the process of becoming successful. This leads to overwhelming bias in Common Wisdom about how things should be done in the workplace that favors large companies and disfavors startups. In other words: most Practices, Policies, Rules of Thumb, Lessons Learned, and so on about how to Get Things Done in the workplace have been written by and for companies that are already well past their startup years. All of this Common Wisdom should be treated with utmost skepticism in the context of a startup since it was written by and large by people whose incentive is to mitigate downside risk. Thus, the higher the upside opportunity / lower the downside risk of your company, the less you should rely on any known Formulas, Best Practices, Rules of Thumb, etc, unless they come directly from successful startups – and even then, the sample size is so small that you shouldn’t really trust those either. There are really only two heuristics that you should rely on because they are self evident: - Startups are successful when they solve a meaningful economic problem faster and more effectively than anybody else does. and - Startups fail when they run out of resources – typically either money, motivation, or both. From these, you can derive a few additional heuristics: - Startups that incorporate market feedback faster are more likely to survive. - Startups that do work that is not immediately relevant to incorporating market feedback are less likely to survive. - Startups that make decisions guided by market feedback are more likely to succeed than startups that make decisions guided by Best Practices, Rules of Thumb, Common Wisdom, or any other mechanism, since those are mere approximations of the market feedback that others received in different circumstances than yours. PART II: Coding Philosophy Since you typically won’t know exactly what code will solve a meaningful economic problem until you’ve written it, your job as a programmer at a pre-PMF startup is to: Write as much code as possible, as fast as possible, to gather as much signal as possible about what the market wants from your company. Everything else is secondary. Code quality; maintainability; reusability; testing; all these are at best auxiliary, and at worst fatal distractions. Remember: Pre-PMF, your company’s upside / downside ratio is somewhere around 95 : 5. You stand to lose ~nothing by taking down Prod for an hour. You stand to gain ~everything by pushing a new feature an hour sooner. Prior to overwhelming PMF, all your code is research code. Your code is the means by which your startup gathers market feedback. The sooner you ship code that provides the market the opportunity to give you information, the sooner you will avoid death by failing to find a solution the market wants. Since all your code is research code, you should be cutting corners everywhere you can manage while trying (but not too hard) to avoid breaking Prod. Refactoring, abstracting, testing, and other Good Coding Practices give you no information about the market’s desires. You should think of them as Last Resorts, when you simply cannot continue to collect market feedback by any other means. Things get a bit more complicated as your team begins to grow. Research spaghetti can be tolerable for the person who wrote it but a nightmare for the next guy. What is the next guy to do? There are two options: Refactor, or Adapt. Refactoring for its own sake is the mind-killer. It is the little-death that brings total obliteration. While refactoring can help with the grokking process for new engineers, it should be only done as part of fixing priority bugs or pushing new features. Refactoring for its own sake – because it’s a Good Practice, or because it may help future engineers onboard, or because it makes you feel better – is a misstep. It prioritizes your feelings over the feelings of the market. The market does not care what the code looks like, but it does care that you aren’t solving its problems fast enough. Far preferable is for new engineers to Adapt. Any engineer with full access to an end-to-end development environment has enough to work with to start pushing features and fixing bugs immediately. While they may feel uncomfortable wading through spaghetti, it is an excellent exercise in getting inside the head of the research coder who came before them. So long as the code they wrote has demonstrated some value in the market, it’s safe to assume that their spaghetti was in fact quite deliberately written despite appearances, and it’s the new engineers’ responsibility to understand why they made the tradeoffs they did. Conclusion Coding at a startup is entirely different from coding for large companies, academic assignments, open-source projects, or hobby projects. Prior to product-market fit, the sole purpose of your code is to gather market feedback as fast as possible. With few exceptions, most Common Wisdom about how to program well or correctly can and should be ignored if your goal is to get rich building a software startup. Shortcuts, workarounds, and kludges are generally preferable to refactors, abstractions and tests as they enable you to gather market feedback more quickly. This post was heavily inspired by @Rengle820, who has led multiple software products from 0 to profitability / acquisition and generously showed me the ropes.
Most engaged tweets of Brian Atwood
My team built the leader competitor to @honey. This "scam" story is absolutely wild to me. In his video today, @MKBHD makes two claims about Honey's foul play: 1. Honey effectively steals commissions from their affiliate partners by being the last to drop their cookie when a user checks out. 2. Honey's value prop to retailers is dubious since they are giving customers discounts on products they were already about to buy. So, why is this wild? Because this is exactly how it has been since the beginning. 1/🧵
It looks like OpenAI is gearing up to release Voice Assistants on Monday. The moment we @SindarinTech have been anticipating for nearly two years. Their quality will almost certainly beat everyone else’s, at least along specific axes. But will they come to single-handedly to dominate the entire market(s) for voice AI? We shall see! Worth noting that we recently converted our default Persona from 3.5-turbo to Llama 3 70b for its superior price / performance and conversational capability.
HOW TO CODE AT A STARTUP An essay about how to think about coding at a startup. PART I: Startup Philosophy The most important thing to remember when working at a startup is the ratio of upside opportunity to downside risk. That ratio should permeate every decision you make. Startups are fundamentally about maximizing upside opportunity, whereas non-startup companies are fundamentally about minimizing downside risk. You join a startup to get rich. Your incentive is to grow the company as fast and as big as possible. You join a mature company to live a nice and/or stable life. Your incentive is to not get fired so you can continue to receive paychecks. When a startup begins, the upside opportunity : downside risk ratio is 100 : 0. When a company has been listed on the S&P 500 for years, the ratio is (usually) ~0 : 100. Prior to (overwhelming) PMF, the ratio is never lower than 90 : 10. Immediately post-PMF, the ratio is somewhere between 80 : 20 and 50 : 50. The company still needs to capture meaningful market share and ensure significant, sustainable margins through further innovation. To reiterate: The upside / downside ratio of your company should guide every decision you make throughout your workday. Much confusion exists about how to get things done at a startup because large, successful companies employ orders of magnitude more people than do startups that are in the process of becoming successful. This leads to overwhelming bias in Common Wisdom about how things should be done in the workplace that favors large companies and disfavors startups. In other words: most Practices, Policies, Rules of Thumb, Lessons Learned, and so on about how to Get Things Done in the workplace have been written by and for companies that are already well past their startup years. All of this Common Wisdom should be treated with utmost skepticism in the context of a startup since it was written by and large by people whose incentive is to mitigate downside risk. Thus, the higher the upside opportunity / lower the downside risk of your company, the less you should rely on any known Formulas, Best Practices, Rules of Thumb, etc, unless they come directly from successful startups – and even then, the sample size is so small that you shouldn’t really trust those either. There are really only two heuristics that you should rely on because they are self evident: - Startups are successful when they solve a meaningful economic problem faster and more effectively than anybody else does. and - Startups fail when they run out of resources – typically either money, motivation, or both. From these, you can derive a few additional heuristics: - Startups that incorporate market feedback faster are more likely to survive. - Startups that do work that is not immediately relevant to incorporating market feedback are less likely to survive. - Startups that make decisions guided by market feedback are more likely to succeed than startups that make decisions guided by Best Practices, Rules of Thumb, Common Wisdom, or any other mechanism, since those are mere approximations of the market feedback that others received in different circumstances than yours. PART II: Coding Philosophy Since you typically won’t know exactly what code will solve a meaningful economic problem until you’ve written it, your job as a programmer at a pre-PMF startup is to: Write as much code as possible, as fast as possible, to gather as much signal as possible about what the market wants from your company. Everything else is secondary. Code quality; maintainability; reusability; testing; all these are at best auxiliary, and at worst fatal distractions. Remember: Pre-PMF, your company’s upside / downside ratio is somewhere around 95 : 5. You stand to lose ~nothing by taking down Prod for an hour. You stand to gain ~everything by pushing a new feature an hour sooner. Prior to overwhelming PMF, all your code is research code. Your code is the means by which your startup gathers market feedback. The sooner you ship code that provides the market the opportunity to give you information, the sooner you will avoid death by failing to find a solution the market wants. Since all your code is research code, you should be cutting corners everywhere you can manage while trying (but not too hard) to avoid breaking Prod. Refactoring, abstracting, testing, and other Good Coding Practices give you no information about the market’s desires. You should think of them as Last Resorts, when you simply cannot continue to collect market feedback by any other means. Things get a bit more complicated as your team begins to grow. Research spaghetti can be tolerable for the person who wrote it but a nightmare for the next guy. What is the next guy to do? There are two options: Refactor, or Adapt. Refactoring for its own sake is the mind-killer. It is the little-death that brings total obliteration. While refactoring can help with the grokking process for new engineers, it should be only done as part of fixing priority bugs or pushing new features. Refactoring for its own sake – because it’s a Good Practice, or because it may help future engineers onboard, or because it makes you feel better – is a misstep. It prioritizes your feelings over the feelings of the market. The market does not care what the code looks like, but it does care that you aren’t solving its problems fast enough. Far preferable is for new engineers to Adapt. Any engineer with full access to an end-to-end development environment has enough to work with to start pushing features and fixing bugs immediately. While they may feel uncomfortable wading through spaghetti, it is an excellent exercise in getting inside the head of the research coder who came before them. So long as the code they wrote has demonstrated some value in the market, it’s safe to assume that their spaghetti was in fact quite deliberately written despite appearances, and it’s the new engineers’ responsibility to understand why they made the tradeoffs they did. Conclusion Coding at a startup is entirely different from coding for large companies, academic assignments, open-source projects, or hobby projects. Prior to product-market fit, the sole purpose of your code is to gather market feedback as fast as possible. With few exceptions, most Common Wisdom about how to program well or correctly can and should be ignored if your goal is to get rich building a software startup. Shortcuts, workarounds, and kludges are generally preferable to refactors, abstractions and tests as they enable you to gather market feedback more quickly. This post was heavily inspired by @Rengle820, who has led multiple software products from 0 to profitability / acquisition and generously showed me the ropes.
Here is the honest reason I haven't raised VC despite having the best product in one of the hottest sectors where my competitors have raised tens of millions: I have zero tolerance for convincing others to believe in me. To a fault. I got into a Harvard PhD with a 3.2 GPA -- maybe a record low -- in large part because I scored 99th %ile on the GRE. Then 1.5 years into my tech career, I was the lead programmer on a product that turned profitable and got acquired (credit to everyone else involved -- the team and my boss @Rengle820 taught me ~everything I know). Then, I married the most beautiful woman I'd ever seen. Now, virtually everyone who tries our product demos agrees that we're the best conversational AI they've ever tried. We'll probably raise anyway. I'm opening up to it. But the words of a much smarter man than I resonate so strongly: "If you don't believe me or don't get it, I don't have time to try to convince you, sorry." So that's why we haven't raised VC.
People with Innovator archetype
Automation specialist
Founder @OTHERSIDEads | Programming the next era of advertising through Connected Performance Ads | Over $500M in ad spend managed | Marketing Enthusiast
I like web 3
Yapper/ Reply guys Feel free to yap with me
19 • cs • ml • space • blogs
founding staff eng @ nuclear startup. private security. neuro, security. Supercomputers.
Architect, Professor & Evil Developer tackling CA’s housing & office mess one Ugly Building & Zoning Hack at a time. Thoughts on CRE, urbanism, design & biz.
swe • learning to build AI to think for me, robots to do my job and a company to fund it all
Building @BerrryComputer @club_homebrew member.
Founder @LudexAI_io | Turning prompts into games with AI 🎮 | Build your first game with just simple words now → ludexai.io
Bitcoiner + game dev. Using AI in a sly, roundabout way to help future generations learn how to fix the money.
Validator/Infra BD @storyprotocol 男的！！！！！
Explore Related Archetypes
If you enjoy the innovator profiles, you might also like these personality types:
