Replacing Engineering Managers with AI Agents

I still remember very clearly the day I talked to a friend about the company I was about to start. He had built and sold several companies and invested heavily in AI before it was cool (again).

I explained to him that I wanted to build a system to help engineering managers regain their time and focus on the important parts of their job instead of the admin. My friend immediately said, why help managers if you could replace them?

 Then he shared this tweet.

The scenario he was describing would look a bit like this: 

Any company could deploy its AI agent, EMAI, to manage its software engineering teams. EMAI is not just any AI. It's been trained on countless software projects from the past decades, holds knowledge from all leading management methodologies, and has real-time data access to the latest technology trends and best practices. It has been specifically designed to replace the role of a software engineering manager.

Let’s indulge this premise and see how this could work in practice:

A Day with EMAI

Morning Stand-up Meetings: Instead of traditional stand-ups, engineers log into their systems and provide a brief update by sending a short voice message. EMAI processes these updates, analyzing voice tones for stress or uncertainty, ensuring it can provide resources or assistance if an engineer faces challenges.

Task Allocation: Using real-time data on each engineer's strengths, past performance, learning curve, and even their preferred working hours, EMAI allocates tasks from the backlog. It uses predictive modeling to optimize for both efficiency and team satisfaction.

Conflict Resolution: If two engineers have a disagreement or are blocked by each other, EMAI steps in. Using its vast knowledge base and understanding of human psychology (aided by its training data), it mediates discussions, ensuring a harmonious team environment.

Training & Upgradation: EMAI monitors the latest tech trends. If a new tool or technology emerges in the market, it identifies which team members would benefit most from training and automatically schedules online courses or tutorials for them.

End-of-Day Reports: Every team member receives a personalized report detailing their accomplishments, areas of improvement, and resources for further learning. These reports aren't just data-driven and include motivational feedback designed to boost morale and foster continuous learning.

EMAI is efficient; it ensures tasks are allocated and problems are resolved with unmatched speed. EMAI is also objective; decisions are data-driven, eliminating potential biases in task allocation, promotions, or conflict resolution.

EMAI also manages to keep stakeholders informed, and it can negotiate with them to find the best solution given their inputs and the business context.

It might sound dystopian, but setting emotions aside and viewing it purely from a business perspective, the idea of replacing engineering managers with AI offers potential efficiencies. In fact, many organizations might be willing to consider this as an experiment. It’s not far off from what a lot of businesses are planning to do or are already doing with customer support departments.

Mind the gap

There’s one thing that’s obviously missing here. The lack of genuine human empathy and intuition might lead to feelings of detachment or isolation among team members. There’s a reason why people just quit their jobs if they feel like all they do is work on an assembly line (no offense here; I spent a few years working in real factories assembling car parts, sandwiches, and more things than I can remember). 

If you step back, however - you’ll notice that ineffective managers can also create a similar environment: rigid processes, constant requests for status updates, and micromanagement turned up to 11. 

Most EMs I’ve talked to have mentioned the fact that they do not want to make the same mistakes their previous managers made, which indicates to me that bad experiences with managers are perhaps the common denominator, especially when you look at how much burnout and lack of motivation is reported within the industry. 

Approximately 80% of developers claim their burnout symptom is a general lack of energy to work and complete coding projects. If empathy is a unique human competence that AI can’t replace, it looks like we are not doing great in the empathy department.

Another argument against AI agents is dependence on data; for example, if there's a scenario outside EMAI's training data, it might struggle to find an optimal solution. However, the same can be said for managers who do not invest continuously in leadership skills or strategic thinking and instead rely on engineering productivity metrics and dashboards charting an average number of comments per pull request (yes, it’s a thing).

And, of course, there are security concerns; relying heavily on AI opens avenues for potential cyber-attacks, but this also applies to any area of business. We're in a new arms race, and new prompt injection techniques are being created every week. AI is here to stay, so we must navigate this anyway.

My friend’s arguments for replacing managers still feel compelling and very plausible, yet I decided to bet on engineering managers powered by coffee (or tea) rather than electricity.

Betting on the future of engineering managers

While AI can recognize patterns and even detect emotions to some degree, genuine empathy comes from shared experiences. Managers can understand personal issues, stresses, and the unique challenges their team members face—both in and out of the workplace. 

Many decisions in the tech world aren't black and white. They require an understanding of various shades of gray born from lived experiences, cultural contexts, and sometimes even gut feelings. Human managers can weigh multiple factors, foresee consequences, and choose paths that might not be immediately evident to an algorithm.

Equally, organizational culture is multifaceted and deeply human. Leaders play a crucial role in instilling values, ensuring a sense of belonging, celebrating successes, and even mourning losses. They shape the team's identity and foster an environment where members collaborate, innovate, and grow. Inspiring people, establishing friendships, and adapting to change are some of the things I can’t imagine being replaced by AI agents (at least for now).

I’m betting on our ability to automate as much as possible the tasks that get in the way of meaningful work so that we can invest in empathy, in being present, and truly serve people in the most humane way possible.