Choosing teams

Company brand matters less than manager + business importance + technical growth over a two-year window. When offers are close on paper, choose the team. You can often transfer teams later inside a large company; you cannot invent a good manager.

Gergely Orosz and others have hammered this for years because it is true: the fastest way to stall is a manager who cannot coach, will not advocate, or runs the team as a ticket factory.

Walking through the three scores

Manager quality. Reverse-interview: who did they last promote and why? Who left and why? How do they run 1:1s?

Business importance. Teams on the critical path get resources, visibility and hard problems. Internal tools teams can be excellent for craft — or quiet dead ends. Ask: if this team disappeared, would the company miss a quarter?

Technical growth. Will you touch production systems that scare you (in a good way)? Distributed failure modes, real performance constraints, security boundaries, ML/inference, payments — pick one frontier. Repeating the same CRUD for three years is comfortable and expensive.

Score the three separately. If any score is poor, demand a compensating excellence elsewhere — or walk.

Talk to two people who left your target company or stage in the last year. Ask what they wish they had known at offer time. Thirty minutes of those calls often beats another hour of Glassdoor archaeology.

Before accepting a high-variance bet (early equity, heavy on-call, relocation), price the downside in writing: "If this fails in 18 months, what do I leave with?" Skills, savings, network and a clear story beat "only stress." If the honest downside is empty, renegotiate scope, cash or role — or decline.

This pairs with company stage: a great manager on a dead-end team at a famous logo still stalls you. A strong manager on a critical path at a growth-stage company is often the compounding seat.

Common mistakes

• Logo first. Manager and path usually eat strategy.
• Skipping reverse-interview. Who they promoted and who left are the data.
• CRUD for three years because it is comfortable. Comfort is expensive.

Check yourself

1. Two offers match on paper. What three scores decide the team?

Manager quality, business importance, and technical growth over about two years.

2. What question probes business importance?

"If this team disappeared, would the company miss a quarter?"

Key takeaways

• Manager, importance and technical growth usually beat logo.
• Reverse-interview the manager; score the three axes separately.
• Price the 18-month downside before a high-variance bet.

Further learning

The Pragmatic Engineer (Gergely Orosz)Observations across the software engineering industry.The Pragmatic Engineer

Next: T-shaped skills.