It’s time to rethink how we measure engineering productivity
The metrics used to measure engineering productivity are obsolete in the AI era. Shifting from velocity to validation fixes roadmaps that keep slipping.
LeadDev
@leaddev
Publisher
19
Posts
Posts by LeadDev
Microservice dogma nearly tanked our seed round
How one startup adopted microservices too early and what reversing that decision taught them about building architecture.
How to hire engineers for strong judgment and taste
Why engineering judgment matters more than ever and how leading companies are redesigning hiring to find engineers who can actually think.
Engineering managers are paying the price of rising expectations
How engineering managers can make leadership capacity visible before judgment, coaching, and technical oversight start to degrade.
AI-coding agents spread through peer pressure, not mandates
Microsoft research on AI-coding agent adoption reveals engineers follow their peers, not mandates.
The 2026 engineer paradox: more capable, but more alone
AI-coding tools are eroding the collaboration and knowledge sharing that made engineering teams stronger – and what leaders can do about it.
Developers on the frontline of the SaaS replacement wave
AI is prompting engineering teams to replace SaaS with custom builds – and the realities that make it harder than it looks.
If anyone can vibe code a product, who needs developers?
Vibe coding can clone your product in days. Here’s what it can’t replicate and why that changes what software developers actually are.
Code maintainability plummets in the AI coding era
New research reveals that AI coding is driving up duplication, suppressing errors, and leaving legacy code untouched.
AI productivity is burning out your best engineers
Why mid-level engineers are burning out under the cost of AI productivity and four changes to make before your best people leave.
Why your AI bill is bigger than it should be
How a $287 AI bill led to an open-source tool that saved users $700,000 – and why token hygiene is the next engineering discipline.
AI-generated code sparks production confidence crisis
Why a third of engineering teams won’t ship their own AI-generated code and what it says about the state of AI coding in 2026.
AI coding is addictive. Engineers are paying the price
AI coding tools are keeping engineers hooked and burning them out. These habits can help before the dopamine loop does lasting damage.
How to prepare for your engineering manager interview
How to prepare for the engineering manager interview – from building a story bank and knowing your leadership style to navigating technical rounds.
AI-coding agents are now the default. What comes next?
How Linear went agent-first overnight, what the productivity numbers look like, and what to manage when AI agents become the default.
Killing a project is every engineering leader’s hardest call
Why defending a struggling project isn’t always good engineering leadership and the rubric that makes the hardest call easier.
Engineering managers ditch cloud AI for local LLMs
Local LLMs are emerging as a third option for engineering managers caught between expensive licences and legal teams that won’t approve cloud AI.
You can vibe code a demo, but what about a product?
What it takes to ship a generative AI product – from inference economics and safety guardrails to the question vibe coding can’t answer.
How to stay technical as a software engineering manager
A practical guide to the four strategies engineering managers can use to maintain technical proficiency.