Software Lead Weekly Software Lead Weekly

How to Be Exceptional at Anything
4 minutes read.

What an incredible post. I've found myself nodding the entire time: "Being exceptional is less about rare talent and more about the choices you make every day. It’s a set of habits that quietly outwork everyone else. Below are five habits for anyone who wants to be exceptional. Show extreme care, for example, if you send a report, open the PDF yourself on another device before you hit send. If you build a product, use it for a week as a user. These micro-actions are cheap, and they compound. [...] A final, blunt point: being exceptional is boring in the short run. It looks like repetition, patience, and low-glory choices. If you want the outcome, accept the work. The payoff is simple: fewer surprises, fewer fixups, and the ability to do bigger work with fewer people. That is where exceptional lives."

Read it later via Instapaper.
Share it via Twitter or email.

Vibe Coding vs. Vibe Debugging
1 minutes read.

My humble effort to help you start the weekend with a smile.

Read it later via Instapaper.
Share it via Twitter or email.

Stop Hiring People
4 minutes read.

"Write a MOC — a document that outlines the mission, outcomes, and competencies of the job. Articulate the business case for hiring, and what, specifically, the new hire should accomplish over the next 6/12/24 months. MOCs are requirement documents for what success looks like. They are for your team what PRDs are for your product. Is a MOC just a JD? No. A MOC should be written before the JD. The MOC is half of what goes into a JD. The MOC’s mission, competencies, and title will show up in the JD. But crucially, the MOC also defines clear business outcomes." -- Funny that after 20 years in the industry, one blog post can hit so hard while feeling so basic.

Read it later via Instapaper.
Share it via Twitter or email.

Vertical AI Is Doing Something Traditional SaaS Never Could: Accessing the 11ドル Trillion US Labor Market Instead of Competing for Scraps of the 450ドル Billion Enterprise Software Spend (Thread)
3 minutes read.

Rohit Mittal writes an excellent analysis that might explain so much of the valuation we see in the market, and how companies grow in markets we used to think were too limited (TAM) or hard to scale: "unlike wrapper startups getting killed every time OpenAI ships a new feature, vertical AI companies build defensibility through domain-specific data, regulatory compliance baked into the product, and deep workflow integration."

Read it later via Instapaper.
Share it via Twitter or email.

Big Ideas 2026
9 minutes read.

Interesting to read the three parts (this is the first one) by a16z's investors, as you can see in which areas they look for opportunities to invest. This can shape how your company thinks about the products you build and the GTM motion to pursue.

Read it later via Instapaper.
Share it via Twitter or email.

More Gears, Same Rider: Using AI Coding Assistants Without Losing Control
6 minutes read.

Great tips to avoid pitfalls when working with coding agents. What would you write and share with your teammates? How can the team build a set of practices and start sharing it with other teams in the organization, where some layers will fit everyone, while others will fit only a few (similar stack?) teams?

Read it later via Instapaper.
Share it via Twitter or email.

Making 20 Upgrades for My Desk Setup (Video)
21 minutes read.

I've been watching Scott Yu-Jan build his workstation, 3D-print, and adjust every small element. Someone send help, I'm worried I'll buy a 3D printer soon that's way too expensive for my needs (or skills).

Read it later via Instapaper.
Share it via Twitter or email.

@SahilBloom: The older I get, the more I realize preparation always beats planning. Planning is based on the expectation of order. Preparation is based on the expectation of chaos. Plan for order and you'll be destroyed by chaos. Prepare for chaos and you'll thrive in any condition.

@Prathkum: Most of coding was never about writing code. AI is just making this more obvious. You no longer need to recall syntax, function structure, boilerplate code, or even API endpoints. That’s the easy part and AI is very good at it. The hard part was never typing. It was always thinking. And it still is.