Coaching Engineers Instead of Managing Them

As engineering organisations grow, the role of an engineering leader has to evolve. Early in a career, engineers often need clear direction: what needs to be built, how it should work, and when it needs to be delivered. As engineers become more experienced, however, continuing to provide that same level of direction can become counterproductive.

Senior engineers rarely need someone to tell them what to do at every step. They need someone who can challenge their thinking, provide context, remove obstacles and help them grow.

This is the difference between managing engineers and coaching them.

Management is often focused on coordination and accountability. Coaching is focused on capability and growth. Good engineering leaders need to do both. The goal is not to stop managing. It is to recognise when an engineer needs direction, when they need support, and when they need to be challenged to find the answer themselves.

Coaching Starts With Questions

One of the simplest changes an engineering leader can make is to ask more questions and give fewer immediate answers.

When an engineer brings a problem, the instinct can be to solve it for them: “You should probably use approach X.” That may solve the immediate problem, but it also removes an opportunity for the engineer to develop their own problem-solving ability.

Instead, a coach might ask:

• What options have you considered?

• What are the trade-offs?

• What would happen if we chose this approach?

• What is the biggest risk?

• What information are you missing?

• How would you explain this decision to the architecture team?

• What would you do if I wasn't available?

• What would you recommend?

These questions force engineers to think rather than simply wait for instructions.

This is particularly important as engineers become more senior. A principal engineer who constantly requires someone else to provide the answer has not really developed the independence expected of the role.

The leader should not always be the smartest person in the room

A coaching mindset requires leaders to become comfortable with not immediately providing the answer. Sometimes the best response to an engineering problem is, “I'm not sure. What do you think?”

That creates space for the engineer to reason through the problem. It also communicates something important: I trust you to think.

Feedback Should Be Continuous

Feedback is one of the most powerful coaching tools, but it is often poorly implemented. In many organisations, feedback becomes something that happens during performance reviews or when something goes wrong.

That is far too late. Effective coaching relies on small, frequent and specific feedback. Good feedback should explain:

• What happened

• What was effective or ineffective

• Why it mattered

• What could be done differently

• What should continue

For example, rather than, “Your communication needs improvement.” A more useful conversation would be:

“In yesterday's architecture discussion, you had a strong technical recommendation, but we lost some of the audience because the explanation started with implementation details. Next time, try starting with the problem and the trade-offs before explaining the solution.”

The second provides something the engineer can actually act upon.

Feedback should not only correct problems

Coaching is not about constantly identifying weaknesses. Positive feedback is equally important.

When an engineer demonstrates strong technical judgment, handles a difficult incident well, mentors someone else or makes a thoughtful architectural decision, those behaviours should be recognised.

If people do not know which behaviours are valued, they cannot intentionally repeat them.

Mentoring Is About Sharing Experience

Coaching and mentoring are related, but they are not exactly the same. Coaching helps someone discover their own answers. Mentoring helps someone benefit from your experience.

An experienced engineering leader has encountered problems that less experienced engineers may be facing for the first time:

• Difficult architectural decisions

• Production incidents

• Failed projects

• Team conflicts

• Technology choices that aged badly

• Successful migrations

• Organisational change

• Career decisions

Sharing these experiences can dramatically accelerate someone else's development. However, mentoring should not become: “Here's what I did, so you should do the same.” The better approach is to ask: “Here's what I experienced. Here's what I learned. How might that apply to your situation?”

Every engineer, team, and context is different. The objective is not to create engineers who behave like their leaders. It is to help engineers develop their own engineering judgment.

Career Conversations Should Be More Than Promotions

Career conversations are another important part of coaching. Unfortunately, career development is often reduced to a single question: “What do I need to do to get promoted?”

Promotion is important, but it is only one part of career development. A good career conversation should explore:

• What does the engineer enjoy?

• What problems do they want to solve?

• What skills do they want to develop?

• What kind of impact do they want to have?

• Do they want to become a technical specialist?

• Are they interested in leadership?

• What experiences are they missing?

• What would they like to be doing in two or three years?

• What opportunities could help them get there?

Not everyone wants to become an engineering manager. Some engineers want to become deep technical specialists. Others want to become architects, technical leaders, product-oriented engineers, or engineering managers.

A good leader does not define success for the engineer. They help the engineer discover and pursue their own definition of success.

Stretch Assignments Create Growth

One of the best ways to develop engineers is to give them opportunities that are slightly beyond their current level of capability. These are stretch assignments.

Examples might include:

• Asking a senior engineer to lead an architecture discussion

• Giving an engineer ownership of a production service

• Asking someone to facilitate a technical workshop

• Allowing an engineer to lead a small migration

• Giving someone responsibility for improving a CI/CD pipeline

• Asking an engineer to mentor a junior colleague

• Involving an engineer in a cross-team initiative

• Allowing someone to present a technical proposal to senior stakeholders

The key is that the assignment should be challenging without being reckless. There is a major difference between: “Here is something you've never done. Figure it out.” and: “I think you're ready to lead this. I'll give you the context and support you need, and we'll check in along the way.” The second creates psychological safety while still providing meaningful challenge.

Give Ownership, Not Just Tasks

Stretch assignments become much more powerful when engineers are given ownership rather than simply additional work.

Assigning a feature gives an engineer a defined piece of work to complete. The technical direction has already been set; their responsibility is to implement it.

Giving ownership of the problem creates a different kind of responsibility. The engineer shapes the technical approach, explores alternatives, recommends a direction, and leads the implementation, rather than simply carrying out a predefined solution.

The second develops capabilities that go beyond coding. The engineer has to:

• Understand the problem

• Explore alternatives

• Make decisions

• Communicate the reasoning

• Manage risks

• Work with others

• Take responsibility for the outcome

These are precisely the capabilities required at more senior levels.

Coaching Requires Trust

None of this works without trust. If a leader asks engineers to make decisions but immediately overrides them, coaching becomes performative.

If leaders delegate responsibility but punish every mistake, engineers will naturally stop taking risks. If engineers are expected to grow but are never given meaningful opportunities, development becomes theoretical.

Coaching therefore requires leaders to create an environment where engineers can:

• Ask questions

• Challenge assumptions

• Make decisions

• Make reasonable mistakes

• Learn from failures

• Seek help

• Experiment

• Take ownership

This does not mean lowering standards. In fact, coaching often raises standards because engineers are given greater responsibility for meeting them.

Know When to Coach and When to Direct

One of the most important skills for an engineering leader is knowing which leadership mode is appropriate.

Sometimes an engineer needs coaching. Sometimes they need mentoring. Sometimes they simply need a clear decision. And sometimes a leader needs to intervene directly. For example, a critical production incident may call for clear direction rather than an open-ended coaching discussion.

An inexperienced engineer may need immediate guidance rather than being left to determine the response under pressure. The immediate priority may be to restore stability by rolling back the deployment, with root-cause investigation following once the incident is contained.

Once the immediate risk is controlled, the coaching conversation can begin. The goal is therefore not to replace management with coaching. It is to develop the judgment to know when to direct, when to support, and when to step back.

Coaching Creates More Independent Engineers

The ultimate measure of successful coaching is not whether engineers enjoy their one-on-one conversations. It is whether they become more capable and independent.

Over time, a well-coached engineer should increasingly be able to:

• Solve problems independently

• Make sound technical decisions

• Explain trade-offs

• Seek appropriate help

• Handle ambiguity

• Take ownership

• Mentor others

• Influence their peers

• Lead technical initiatives

The leader's role gradually changes. Initially, the leader provides clear direction and guidance. As the engineer grows, the relationship becomes more collaborative, with greater shared discussion and independent decision-making. Eventually, the engineer begins coaching and mentoring others, passing on what they have learned and helping engineering capability compound across the organisation.

Coaching Is a Force Multiplier

An engineering leader can only directly solve a limited number of problems. A leader who develops ten engineers who can independently solve problems has dramatically increased their impact.

Those engineers then develop others. This creates a multiplier effect:

Coach → Engineer → Capability → More Engineers → Greater Organisational Capability

This is why coaching should not be viewed as something that takes time away from engineering delivery. It is part of engineering delivery. Every hour spent helping an engineer develop better judgment, communication, technical capability or leadership skills can reduce future dependency on the leader.

Coaching in an AI-Driven Engineering World

AI makes coaching even more important. Engineers now have access to tools that can generate code, tests, documentation, designs, and even proposed solutions in seconds.

This changes what engineers need from their leaders. The emphasis shifts from producing code to developing the judgment to assess whether it is correct, appropriate, secure, and maintainable.

Leaders therefore need to coach engineers to:

• Challenge AI-generated solutions

• Validate assumptions

• Understand trade-offs

• Review generated code critically

• Identify hallucinations and incorrect reasoning

• Protect sensitive information

• Understand when AI should and should not be used

• Remain accountable for the resulting artefact

AI may reduce the amount of work required to produce software, but it does not remove the need for engineering judgment. In some ways, it makes that judgment even more valuable.

Conclusion: From Managing Work to Developing People

Coaching shifts the focus from completing work to using that work as an opportunity for growth. Leaders give engineers space to solve problems, use questions to develop their thinking, provide continuous feedback, support meaningful career choices, and offer carefully chosen stretch assignments.

The aim is to help engineers become more capable, confident, and independent, not dependent on a leader’s answers, but able to find answers themselves.