Ash Rust

Anthropic Tries to Save America This week, the US Military demanded unfettered access to Anthropic's Claude, which the Pentagon already uses. Specifically, the Military wants to remove 2 safeguards surrounding mass domestic surveillance and autonomous kill decisions. To their credit, Anthropic said no. Honestly, these don't seem like strong enough safeguards, for example there's nothing in those two limitations regarding subverting elections. The only credible argument for removing these restrictions is that our enemies won't hesitate to use these new tools for all this and more, putting us at a dangerous disadvantage. Of course, those are the exact same arguments made for torture, war crimes and chemical weapons. In America, our strength comes from (trying) to choose the right path even when it's difficult, with both our domestic freedoms and how we treat others. The example is a beacon, that has attracted great people to this country from all over the world, since its founding. Some might claim America has lost her way of late, but a San Francisco-born CEO risking a 9 figure contract to stand up to their government in the name of Freedom... that's exactly the kind of patriotism and strength we can all still be proud of.

Try to focus on a narrow customer profile early on, it’s much easier to learn what they want.

The Top 3 Reasons We Don't Invest We meet hundreds of teams each year and invest in only 15–20. However, clear patterns exist in the reasons why we choose not to invest in a startup. Here are the most common reasons why we pass and what we're looking for: 1. Tech Advantage. At the earliest stages we don't expect you to have already built world beating technology, but that should be the plan and you need the right team to do it. If your goal is to build an interface improvement or something else others can copy, that will be a blocker. 2. Equity Split. We prefer a roughly even equity split among the founders, considering most of the business still needs to be built. Even if one person had the idea or started earlier. If one founder is investing, treat that like any other investment. Heavily skewed splits usually concentrate decision-making, and the spoils, making it harder to retain the other co-founders long term. 3. Suites. Focus is a big part of startup success, so we want teams to concentrate on one feature versus several. The logic is that focusing on one narrow feature will make it hard for others to compete, potentially forcing them to rely on your technology. Although these are our preferences, many successful businesses took a different path. If our approach is right for you, we'd love to chat: sterlingroad.com/apply

How to Get More Investor Intros New investors are one of the best sources of introductions during fundraising. So if you’re not maximizing your intros from investors, you’re slowing the round. Here’s how to do it: 1. Ask them. When you confirm receipt of the wire ask the investor who else they would suggest. This will usually be a short list of their friends and people they spoke to recently. 2. List Review. You should have a target list of investors you want to reach. Show this to your new investor, asking for more intros. 3. Network Search. Use tools like Happenstance or Sales Navigator to find other investors in their online network. There will be plenty of people you missed from steps 1 and 2.

Great to be part of @ImperialCollege IdeaForge again this year. Maksim and the team have built a strong community of founders. We are proud to sponsor the £3,000 prize. Congrats to Ossava on the win!

15 meetings a week is the baseline in fundraising. 100 meetings in 6 weeks, and most talented people can pull a round together.

@kayintveen @SandervHooft @softwincn @aquavoice I haven't touched an IDE in 6 months. I use textedit/notepad as a scratch pad for token copying etc. Otherwise 100% Claude Code.

@AshRust @SandervHooft @softwincn @aquavoice yep docker desktop removes the last real platform dependency for the app side. what's left is just IDE muscle memory and keybindings. technical problem is already solved, its all habit at that point

20 years ago I moved to Mac and I’ve used one ever since. Last week, I went back to Windows because of Claude Code… This is a @softwincn Pocket 4, it’s insanely small with an 8.8 inch screen and full laptop functionality. I have a small usb mic (right side) and use @aquavoice to do most of the typing. It’s working great! @Apple has been stagnant for a while on their laptops, so I just couldn’t buy another. Wish me luck…

@kayintveen @SandervHooft @softwincn @aquavoice Docker desktop is on both mac and windows

@AshRust @SandervHooft @softwincn @aquavoice docker solves the deploy side completely. the friction that stays is the local dev environment - IDE, terminal, local tooling config. thats still surprisingly platform-sticky. curious how that side has been on windows so far

@kayintveen @SandervHooft @softwincn @aquavoice I use docker to deploy…. So no issues moving between machines even if local.

@SandervHooft @AshRust @softwincn @aquavoice thats where its heading. remote containers, browser IDEs, AI running in the cloud. the local OS increasingly becomes a display layer. still on mac and not reversing that anytime soon but the stickiness is definitely lower than it was

You can rely on paid ads to acquire new customers. Lots of companies use ads successfully. But keep in mind that ads are high risk because, should you ever have budget problems, and most companies do, the reliance on ads can become a death spiral. 💀 Listen to the full episode of The Startup Help Desk available wherever you get your podcasts. 🎧

@Nash @softwincn @aquavoice $1300 or so on amazon

@AshRust @softwincn @aquavoice How much was it for?

@IAmMattGreen @softwincn @aquavoice Yes same size an ipad mini

@AshRust @softwincn @aquavoice what does this do that apple laptops don't? Is it just that it's really small/portable?

When it comes to fundraising, momentum makes all the difference ⬇️

@toddgoldberg @zachcoelius @Winterrose It is time.

@zachcoelius @Winterrose @AshRust Indeed. He’s finally dragging me out to one in a few weeks.

are there any places to smoke cigars and shoot guns in SF with cool people? realizing that sounds fun

How to Cold Email an Investor Even before mainstream AI, cold emailing investors was a low probability gambit but nowadays it's truly lottery ticket odds. For example, over the last year, I've received ~10k cold emails but I only replied to 3. If you really have no other way to reach an investor, here's the process: 1. Check contact preferences. Many investors have application forms or preferred modes of communication, if you don't follow their request, you put yourself at an immediate disadvantage. For example, Sterling Road has an application form on our website and my LinkedIn profile asks visitors not to use DMs. 2. Strong Subject. Grab their attention, by highlighting the company's biggest strength in the Subject. As an idea stage investor, team matters most to me and the 3 I replied to all mentioned team strength in the Subject. Other examples: "We scaled Acme to $50M ARR", "Stripe/Oxford Team starting something new", "Our Research was Featured in Nature". 3. 3 Key Points. Once the investor opens the email, you have only seconds to maintain their interest. State the company's progress clearly in 3 concise bullet points, typically covering traction, technology, and team. For example: "We worked together at OpenAI for the last 2 years", "We have 50 customers on our waitlist and 10 have paid >$1k", "Users spend 3x longer on our app versus Google". 4. Under 100 Words. Keep the message incredibly short. Since you already have 3 bullet points and will need a polite closing with a meeting link, you essentially only have room for your one liner, that's it. Email attention spans are very short, every word over 100 reduces your chances of a reply. Cold emailing investors is a channel in decline and you should always try to find another route first. However, if you've exhausted all the other options, following these rules will help you get a response.

@garrytan I’ve used Eng plan review a lot today

My CTO friend texted me: "Your gstack is crazy. This is like god mode. Your eng review discovered a subtle cross site scripting attack that I don't even think my team is aware of. I will make a bet that over 90% of new repos from today forward will use gstack."

The 1st part of distribution is answering these 3 questions, don’t try to scale immediately.

The AI Tools Used by Sterling Road Founders I'm often asked what tools I recommend for getting started building with AI, so here's what Sterling Road portfolio founders use. 1. Getting Started. If you're prototyping or building with AI for the first time, Replit is a great place to start. 2. Coding Tools. For bigger projects, most Sterling Road founders use Claude Code with some using ChatGPT Codex for debugging. Cursor is also used but that has declined recently. 3. Deployment. Vercel is by far the most popular option, though AWS and Google Cloud are sometimes mentioned.

@garyautomates @maticrobots yes

@AshRust @maticrobots Can you create rooms and have it clean one area on a schedule or on demand?

Last week, we got our first robot in the home. A @maticrobots vacuum cleaner & mop. I was very skeptical but this thing is actually impressive. 1. It avoids obstacles very intelligently and maneuvers well. It’s only gotten stuck once. 2. It does a surprisingly good job of vacuuming, even handling our numerous rugs. 3. Privacy - it does not need a WiFi connection to work. We use ours on Bluetooth only. Overall, very happy. Not an investor but I wish I was.

You don’t need many materials to raise a seed round but they need to be good: