Everything you’ve been told about burnout is wrong

Everything you’ve been told about burnout is wrong
3 minutes by Blake Scholl

Burnout is often misread as doing too much for too long. In reality, it hits when the next meaningful goal feels too far away or out of reach. Breaking a big goal into smaller, regular milestones gives people something believable to work toward. When motivation is low, Blake suggests checking the goal before checking the workload.

Don’t sacrifice quality for speed (ElevenLabs definitely doesn’t).
sponsored by Meticulous

Agents are writing code faster than ever. They’ve even been entrusted to review their own work, but who’s reviewing them? One does a passable job and ships a buggy fix, the next copies it, and bugs spread faster than anyone can catch them. Leadership at Dropbox, Notion & ElevenLabs refused to trade this quality for speed. With Meticulous, every change is exhaustively verified, with zero developer effort.

Change creates opportunity. It also sucks.
12 minutes by Molly Graham

AI is changing jobs quickly, making some skills less valuable while creating new opportunities. Molly argues that adapting means letting go of old responsibilities, accepting the discomfort of becoming a beginner, and focusing on what comes next. However, not everything should be handed to AI. Human judgment, creativity, relationships, accountability, and unique strengths still matter. The goal is to use AI for leverage, not replace everything humans do.

Why good product ideas don’t survive
13 minutes by Itamar Gilad

Most product ideas deliver little value, and truly great ones are rare. Traditional development processes make this worse by killing good ideas through political prioritization, committee-driven design decisions, and rushed delivery. Testing ideas with real evidence before building them in full dramatically improves survival rates, filtering out bad ideas while protecting the good ones.

How to build innovation teams
20 minutes by Phil McKinney

Hiring smart people and giving them space isn't enough to build an innovation team. The people who will innovate are already doing it, and the job is to find them and clear the way. Phil shares six steps that matter, in order: build a culture of trust and permission to fail, recruit for four specific roles rather than just talent, adapt your process instead of copying another company's, fund ideas outside normal budget cycles, measure what shows leadership real value, and lead the work directly every day. Get the order wrong, or skip a step, and even talented teams fail.

Evals as theory building
12 minutes by Alvaro Videla

Evals shape what AI systems learn to optimize, so a flawed eval quietly steers everything in the wrong direction. A score can rise because a system genuinely improved, or because it learned to game the measure. ProofPack pressures evals with known failures and controlled counterfeits, then keeps the full evidence trail so results can be checked by others. Assay locks the decision rules before results arrive, so conclusions stay honest.

And the most popular article from the last issue was:

• I don't want the details