Developers must now direct AI tools, evaluate their output, communicate tradeoffs, and make sound technical decisions rather than simply writing every line of code.
Preparation starts now. Focus on these three areas.
Tip #1: Learn to direct AI, not just use it
Execution looks different now. Great work means defining the problem clearly, providing context, reviewing generated code, and deciding what ships. As AI agents handle more implementation, these skills become more valuable.
Consider adding a new authentication flow.
A traditional workflow follows this pattern:
Task: Add authentication
Create branch
Write code
Run tests
Open pull request
As AI tools integrate deeper, these same skills help coordinate multiple agents. Your workflow shifts to this model:
Workspace: Add authentication
Agent 1: Authentication ready for review
Agent 2: Documentation draft ready
Agent 3: Test suite ready
You remain responsible for the outcome. You spend less time implementing every piece. Instead, you define the work, review outputs, and make the technical decisions that bring everything together.
Learn to direct AI agents.
Tip #2: Don’t trust AI’s first answer
AI generates solutions in seconds, but the first answer is not always the best. Your experience writing clean, maintainable code helps evaluate output.
Ask a second AI model to critique the first model’s work. Then use your own judgment to evaluate both responses.
Here is a prompt showing the process:
“Write a SQL query that returns each customer’s most recent order.”
AI Model #1 generates the query.
AI Model #2 (Critique) flags issues:
Does not handle duplicate timestamps
Missing index recommendation
May perform poorly on large tables
Different AI models have different strengths and blind spots. That is why GitHub Copilot’s built-in Rubber Duck agent uses a second model to critique plans, code, and tests before you move forward. A second perspective catches issues the first model misses.
Trust AI enough to use it but not enough to skip review.
Tip #3: Use AI to solve bigger problems
When AI saves implementation time, developers use that time to tackle broader problems. These include understanding customer needs, evaluating tradeoffs, designing better systems, and making technical decisions that AI cannot make.
Consider Issue #4821 to add dark mode.
AI handles the build implementation, generates tests, and updates documentation.
The developer checklist remains:
Validate customer problem
Review architectural tradeoffs
Check accessibility
Define success metrics
Approve solution
As AI takes on more implementation work, the skills that distinguish great engineers become more important. These include exercising sound judgment, balancing tradeoffs, and solving the right problems.
Let AI handle more of the implementation so you can spend more time building the judgment that helps teams make better decisions.
What it means
The shift moves the developer role from manual coding to system oversight. You define requirements, verify results, and make decisions on architecture and tradeoffs.
