Solve the hard problem first
Successful leaders attack the hard problems first, because those are the problems when solved, unlock the entire equation.
Two truths and a lie about software engineering
The carrying costs of production software are recurring. Sadly, we rarely realize it, much to the dismay of stakeholders. Having a healthy appreciation for the carrying costs of code will make you a more effective leader.
The first rule of decision-making
Leadership requires making decisions and I've found it helpful to apply three principles.
Extending Code Llama
Retrieval-augmented generation (or RAG) boosts the usefulness of AI by combining specific knowledge with an LLM. In this case, pointing Code Llama at specific code yields a highly productive learning experience.
Pairing with my good friend Code Llama
AI won't take your job, but someone who's leveraging AI will. Besides, it's easy getting started with AI, especially if you're looking for a coding buddy.
Perverse incentives and DORA metrics
“Move fast and break things” isn’t what anyone wants. Plan for complex systems to break. Reframe the rallying cry to “move fast with reliability” and stop wasting time measuring failure.
The value of values
Values represent what you consider important and desirable in life. If you want to find meaning and purpose, then know your values and be true to them.
Supercharging PostgreSQL with PgBouncer
PgBouncer enables you to both vertically scale Postgres and more efficiently take advantage of Postgres horizontal scaling.
Mining for dissent
It feels good to be in a room of yes-people, but it’s a trap. Be brave and seek out if and why people disagree with you. It'll make you a more effective leader.
