Jeff Shafferman

You don't need to hire a software farm for anything anymore. You need one developer who sits next to you for 6 weeks and learns how your business runs. Not how it looks in a deck. How it actually runs. The messy inbox system. The Notion doc nobody follows. The 4-step process your EA does manually every morning. A good AI developer, someone who works in n8n, Claude, and modern automation tools, turns that into systems that run without you. We placed one with a 12-person e-commerce brand. In 8 weeks he'd automated their order exception workflow, their influencer outreach follow-up, and their weekly reporting. The admin team is being trained in AI. Those that don’t adopt are gone. 3 tasks. Each one was eating 6-8 hours a week from someone on the team. That’s money in your pocket. 416h per year. That’s almost a full quarter's worth of work. A dev shop's savings don't matter if they can't ship. Ours do. I wrote more on this on more detail in the blog linked below: https://lnkd.in/gVpWJGZw DM me if you have been seeing AI happening in front of you and want to know how you can avoid missing out.

32

10 Comments

When a dev leads with “clean architecture,” I pause. If they push for the perfect setup before writing a single feature, I ask why? It’s not that quality doesn’t matter. It’s that context does. Perfect code doesn’t help if no one needs what you built.

7

4 Comments

I asked AI to generate an image of what it’s like chatting with me every day. It came back with: “This is fine. Everything is fine.” 😅 If you’ve ever shipped software in a real enterprise, you know the feeling: * the pipeline is green… except in QA * the “small change” touches 6 services and 2 vendors * the architecture diagram is clean… until you zoom in * someone’s “source of truth” is a spreadsheet from 2017 But here’s what I actually like about AI in the middle of that chaos: Not that it writes code. That it helps me keep momentum when the system is loud. It’s a second brain for the unglamorous work: * clarifying the end-state before we build the wrong thing perfectly * turning tribal knowledge into repeatable patterns * tightening contracts and boundaries * translating complexity into plain language so decisions can happen “This is fine” isn’t denial. Sometimes it’s the moment right before you turn the mess into a plan. #engineeringleadership #softwarearchitecture #platformengineering #devops #ai

8

Hot take: Developers who marry themselves to one tech stack are limiting their own growth 🚀 After 8 years deep in .NET, I've realized something—switching languages/frameworks isn't the mountain we make it out to be. Give me 2-3 weeks, and I'm production-ready in pretty much anything. Why? Because the fundamentals don't change. Problem-solving, architecture, clean code—that's the real skill. The syntax? That's just details. Here's what bugs me though: Companies are still hiring like it's 2010. "Must have 5+ years in [insert specific framework]" Really? In the AI era where tech evolves every quarter? What if we flipped the script: - Hire for problem-solving ability, not tech stack memorization - Embrace multiple stacks for different problems (right tool for the job) - Consider part-time specialists over full-time single-stack devs - Build more resilient products with diverse tech options The math actually works out—more cost-effective, stronger products, better team flexibility. Am I crazy here, or are we overdue for a hiring philosophy upgrade? Drop your thoughts below 👇 #SoftwareDevelopment #TechCareers #DeveloperLife #CodingCommunity

14

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?

45

16 Comments

They were whispering the answers off camera?!?! Let me explain... When selecting a software development sub-contractor or "development partner", how do you really know if they're any good? When hiring a developer as an employee, there will usually be a hands on element of the interview. Using this approach I've built exceptional in-house teams for clients, with resulting teams that benchmark at the top of the scale, let alone top of the industry (with the DORA Software Delivery Performance '4 keys' metrics – see: https://lnkd.in/eSf8zeAW ). Do you do the same with your suppliers/partners? If you don't, the resulting technical debt you inherit might not be apparent until long after the last of their invoices are paid and your in-house teams take over (seen this so many times)! Years ago, a client asked us to do such an evaluation. The supplier (a big name) was asked by our client to arrange for us to pair with the people that the supplier had in mind for the engagement. The results were shocking... Someone could be heard whispering instructions in the background! Even those instructions fell short of even basic competency. The consultants they had in mind were barely "advanced beginners" and certainly fell short of "competent" (see the Dreyfus Model)... And this was despite the supplier's sales people having talked a good talk, claiming they had the best people in the business. Suffice to say, the client went elsewhere. That was a smaller engagement... For larger engagements, I would start with a DORA 4 keys evaluation of a sample of the supplier's teams, which I'd also ask to see evidenced. Then, I'd select a random sample of those teams to mob-program with to reinforce that evidence. Only then could I endorse that supplier... So, when selecting your subcontractor/partner, just keep in mind that the firms with the best sales people with the biggest budgets to wine and dine you sometimes have the worst consultants... And sometimes the companies with the worst sales people, who sell on their people's merits and not expensive meals, have the best consultants. Navigating perceptions vs reality is the key. As they say, never judge a book by it's cover.

1

🚨 Game-Changer in Test Automation Just Dropped! 🚨 Jason Huggins — the creator of Selenium — has launched Vibium, and it’s not just another framework. It’s a game-changer in stable, scalable automation. Why Vibium stands out from existing tools: ✅ Self-healing test selectors → no more failures due to changing locators ✅ AI-driven UI adaptation → dynamically handles app changes ✅ Unified automation platform → Web, Mobile, and API in one place ✅ Cloud-native execution → massive parallel runs with minimal setup ✅ CI/CD-ready design → reliable, fast, and less flaky ✅ Write tests in plain English → no coding expertise required In short: Selenium gave us automation. Vibium gives us automation that actually works — consistently, at scale. 💡 Documentation coming soon—stay tuned! Fewer flaky tests. Less maintenance. Faster pipelines. Vibium could well become the new standard in test automation. #Vibium #TestAutomation #QA #Selenium #JasonHuggins #Stability #FutureOfTesting

3

Has anyone seen a plugin for slack that would do automatic message digest with a breakthrough code? After reading "Deep Work" at a coworker's recommendation, I've altered my work to be a couple of focus blocks during the day during which I don't communicate. Then I have about an hour of following up on messages and occasionally emails. This has been a huge productivity boost for me with almost no perceivable loss of important response. The ideal would be that messages (chat, email, etc.) would accumulate in a bucket during my focus block. At the end of the block I would empty the bucket and everything becomes either a response, a task, or gets archived. Obviously I'd want a breakthrough feature that people could use if I'm in a focus block and the thing they need from me is time sensitive. But it could then be a an opt-in of "Peter will see your message in 1h 23 minutes. If he needs to see it before then, click here". I'd definitely pay to have this feature.

2