Eli P.

Openclaw joining OpenAI feels like the closest signal yet that the “one-person unicorn” era might not be fiction. If this model scales, headcount stops signaling capability, leverage shifts to agentic systems and architecture, and founders win on system design—not org size. We’re moving from “build a team” to “build a machine.” #AI #Startups #Founders #AgenticAI #FutureOfWork

2

3 Comments

If you’re a US startup hiring mobile this quarter: screen for release ownership, not Flutter trivia. Senior candidates don’t separate “engineering” from “shipping.” They own releases, hotfixes, and the messy production stuff that actually matters. One question that works: “Tell me about a release that went sideways. What broke, how did you diagnose it, and what did you change so it wouldn’t happen again?” The answer tells you if they’ve actually owned production when it broke.

17

5 Comments

Why neutrality in #SoftFurnishings is harder than boldness (And why I’ll always choose it.) 1️⃣ Neutral isn’t minimal. People often confuse neutral palettes with playing safe. But designing a neutral space that feels complete takes far more restraint than going bold. It’s easy to fill a room with colours and patterns. Much harder to know when to stop. 2️⃣ It takes discipline to edit. When clients see an empty corner, their first instinct is: “Let’s add something.” My job is to pause, feel the energy, and ask, “Does the space need more? Or are we avoiding stillness?” #HealingSpaces often breathe through the things we remove. 3️⃣ Quiet choices hold deeper power. A soft textured rug or a muted sheer aren’t background details. They are anchors. They guide the way you feel, move, and think. 4️⃣ Neutrality requires deep listening. Not just to the client, but also to the space. The walls. The light. The emotional weight of what already exists. If you listen carefully, you’ll often realise: The space always says something. You just need to honour it. 5️⃣ Bold is easy, but balance is tricky. Using statement colours is easy.  It takes a calibrated mind to design a space that feels peaceful and subtle. That is what makes neutral spaces feel timeless. Not boring, but deeply human. 💬 Have you ever walked into a room and felt instantly calm, without knowing why? *********** 🏡 Follow Darsha Manek for more on Soft Furnishings. *********** #LuxuryDesign #Entrepreneurship #EnergeticAlignment #Fabrics #Trends #HealingSpaces #SoftFurnishing

13

A team member messaged me after a 4-hour call fixing dev issues. "I can't believe you were on that call." We have 100+ people at Genflow. I run a $100M agency. And I was right there debugging code with the team. It hit me how wrong most people get leadership. Everyone thinks being CEO means strategy sessions and vision boards. Wrong. Being CEO means being the captain who's still on the field. You show people how it's done by doing it alongside them. Not above them. The idea that you'll build a business and just work on "high-level stuff" while everyone else executes? That's not how you scale. You scale by staying close to the work. Understanding the actual problems. Building alongside your team. After 9 years of doing this, I've learned: The best CEOs aren't in corner offices. They're in the trenches when things break. They know the product inside out. They can still do the work if needed. Your team doesn't follow titles. They follow leaders who roll up their sleeves. What kind of leader are you trying to be?

44

16 Comments

What is this stack overflow? Anyone got good examples of parsing work summaries for the subject parts? Assuming no other context provided (aside from the summary and potential mappings), not even the source code changed. Example: Software Engineer or AI summarizes and writes "Fixed bug relating to Status API overages. Removed useEffect in favor of mobx" I assume the prompt would -parse text for objects -assume speaker is acting on and is the subject -provide or allow for context that maps the objects discovered to internal business metrics/surface areas in the example above I would want structured output that tells me -application areas (front end, status page) -business areas impacted (API costs) -components touched (client side api, react) i should post on stack overflow.

2

Several people know about this problem but nobody talks about how to solve it. I was asked one time to help fix this issue at a company. The fix lies in breaking the work down to a few critical steps. First, never just hand off work in its entirety and expect you'll get anything you can use. Regardless of how good or bad the team is, always ask, always, the offshore team to pseudo code their approach. They should explain which files they'll touch, which systems they'll touch, integrations, which areas or layers of the application, etc. I call this the first pass. Why is this important? This tells me they're thinking about the problem right and if there are any gaps in their understanding, or any pieces they're missing. After that, they'll do a second pass. That is stubbing out which functions or code will need to be added or modified, maybe even making the functions and service calls work but with mock data. This is the "prove it to me" moment. This second pass also helps me as a dev manager gauge how fast or slow the offshore team is working, giving me indications too if they're stuck somewhere. Only after this second pass, do I actually turn them loose to go execute and "deliver" the solution. Only now, because I did some diligence on my managerial end, I've got greater confidence that what I'll get will be something on par with what I'm expecting and can use. This partnership and collaborative arrangement also fosters trust and over time the offshore team will learn how your organization operates, critical thinking, and that nothing should ever be done in a vacuum or cowboy way. For more helpful tips like this, follow me. I'm paying it forward in 2026 and will be sharing gems most people aren't talk about with a focus on the solutions, not just problems.

1

Developers weren’t just builders. They were gatekeepers. If you couldn’t code, you didn’t create. You waited. On engineering bandwidth. On roadmaps. On someone else’s priorities. Syntax was the wall. Know where the semicolon went? You shipped. Didn’t? You stood outside. That divide shaped more than workflows. It shaped status. Developers got the leverage. The scarcity gave them the power to decide what shipped and when. Everyone else adapted around them. It wasn’t about being smarter. It was about access — and access wasn’t evenly distributed. Some people were in the right rooms early, tripped into syntax, and built from there. Others never got the chance. That was the difference. But syntax was only the first kind of alpha. The second was quieter. Harder to see. More enduring. System knowledge. Knowing which database could survive Black Friday traffic. When to debounce vs. throttle. How to shard writes across regions. Not just writing code — architecting it. That’s the tacit layer. Built from war stories. Outages. Scaling pain. Years in the trenches. It’s still here. And it still matters. For now. I remember the first time I tried to shortcut it. I’d tell Claude what I wanted. It would spit out code. I’d paste it. It would break. I’d send back the error. It would try again. Loop by loop, something shifted. Reddit called it “copy-paste monkey.” They saw a hack. I saw a door opening. People who’d been locked out started building demos. Then products. Then real companies. The wall between “technical” and “non-technical” was never about intelligence. It was about translation. And now, translation is free. What matters now isn’t syntax. It’s clarity. How well you can think. Break things down. Iterate. That’s what the agent understands. And that’s what it rewards. The syntax advantage is gone. Developer privilege is flattening. System knowledge is next. Today, the agent can’t always choose the perfect database. But if you tell it the trade-offs, it will. If you give it context, it will architect with precision. That tacit edge still holds. But only for now. Soon, even that will be embedded. And leverage will shift again — to anyone who can think clearly enough to guide the machine. That shift isn’t coming. It’s here. And it’s not going away.

18

2 Comments

You'll be shocked to know I have a contrarian take on prompt design for writing code. I'm seeing a consensus form around very explicit, comprehensive instructions in the prompt. Several people have sent me their prompts asking for help and I'm always surprised at their level of detail. I can only link to my chat, because it's long and meandering. It's worth thinking about how these models are created. Initially, they are trained on the next most likely token in a bunch of found material. That means if you want a web app you need to write the header for a web app so it can complete it. When LLMs we're in this pre-chat phase in the late 2010s, they were interesting but not useful. Chat models are fine-tuned after this initial training with conversational data so they learn how to build your web app in response to an ask rather than the beginnings of a file. So modern LLMs are designed for conversation, not for a massive requirements dump that leads to a completely clean result. Reed Hastings says to lead with context, not constraint. This is good advice not only for people but also for LLMs. In fact, wanting one shot answers from a complex prompt is probably the biggest source of mismatched expectations for these models. People who don't see the value are usually frustrated with the non-determinism and uneven results. Change the prompt slightly and you get a totally different implementation. Others prompt "come up with five implementations and then pick the best one". Or they set up an agent to do it 1000 times. I do, too. When I prompt, I use a conversational approach. I never run a slightly modified version of a prompt a second time. I just fail forward. When I have wanted a repeatable prompt---because I'm creating automated API access---I have ChatGPT write the prompt for me, and I iterate on it conversationally. When writing a requirements document to start a new code project, I have a discussion until I'm confident the context captures what I want. Then I ask the model to write a requirements document for implementation. If we've been talking for a while, I rarely see this stage introduce oddities. And if the code doesn't work, I iterate and use the requirements document to ask where the code is deficient and then (after it responds conversationally) to fix the problems it just described, assuming I didn't need to refine them first. You just need to drink the Kool Aid. Treat it like people. If your junior engineer keeps writing crappy code, you don't go back and tweak the requirements document and hand it to them again. You also don't go and silently change their code line by line. You point out issues and ask questions at a high level in a conversation. You give them context, not constraint. Then I go back to this platform and I see everyone's advice is engineer your prompts more carefully and provide more constraint. Am I the only one who finds stream of consciousness conversations with TMI beats engineered prompts?

41

46 Comments