Intention vs Mechanism for Team Work

Sharing one of the best lessons I've learned from my onboarding plan two years ago.

Robotics - A high-level view [part 1]

Continuing my reflection on the conversations at GHC 2024, I'd like to share another popular topic - my high-level overview of the field of robotics.

Robotics is a multidisciplinary field. Nowadays, when people hear the term "robotics," the first things they might think of are “AI,” “machine learning (ML),” or even “large language models (LLMs).” While these technologies are starting to have an impact on the field, robotics encompasses far more.

Fundamentally, robotics is a field that requires both hardware and software.

Hardware
Robots are designed to interact with the world, often physically. This means the robot needs to have “links” that form the robot's body, “actuators” that provide motion, and “sensors” that can observe the environment. Each of these components is a research field in itself, although the innovations may not be making headlines as often as AI.

• Links: I use the term “link” in a general sense. Links connect actuators. Subjects such as materials (e.g., aluminum tubes vs. carbon fiber tubes) and connectors (e.g., different ways to “lock and unlock” two parts) are still very active research topics.

• Actuators: This includes traditional motor-driven actuators, pneumatic actuators, and various novel soft actuators. Another significant topic is “gearboxes.” At a high level, a gearbox is used to convert angular velocity to torque and vice versa. (This is not magic because "power" remains the same in this process. In fact, there is often a cost, i.e., transmission loss. This means when we use a gearbox to convert between angular velocity and torque, we lose some power due to mostly friction.) Here are some videos I found inspiring just by looking at them (I guess this shows my mechanical engineering background :)

• Sensors: These include parts such as cameras, lidar, and many more. Consider how the cameras on smartphones have advanced. These improvements continuously expand the range of applications that robots can perform.

Next time we will talk about software!

Industry - Intention vs mechanism

This week has not been the easiest at work. I'm lucky to have a group of teammates to talk through the challenges we're facing. Even though discussing it helps emotionally, we still need to fix the problem or at least improve the situation. This situation reminds me of a concept I learned during my onboarding training when I first joined as a full-time employee - Mechanisms at Amazon.

The training started with a quote from Jeff Bezos: “Often, when we find a recurring problem, something that happens over and over again, we pull the team together, ask them to try harder, do better - essentially, we ask for good intentions. This rarely works... When you are asking for good intentions, you are not asking for a change... because people already had good intentions. But if good intentions don't work, what does? Mechanisms work.”

This was a game-changing moment for my perspective on teamwork. As a person who always wants to become a better version of myself, I often project the same mindset onto others. This makes it natural to think that “problems can be fixed if people improve.” The actionable item then becomes “how can we help people be better equipped or motivated?” This way of thinking is reasonable, but as Bezos pointed out, it rarely works if the goal is to fix the problem.

On the other hand, Mechanisms are complete processes that include tools, adoption of the tools, and inspection to ensure the intended outcome is achieved. Improving the mechanisms helps us focus on the problem, not the people involved. (Of course, we want to have awareness of people and their personal development. The main thing is, a fix to the problem should not depend on people's development.)

I find this "mechanism" approach helpful. First, it helps me find more efficient and direct solutions to problems. Second, it prevents me from questioning teammates and thinking, "Is XYZ working hard enough?" Note that this is not to say people are not key to solving a problem. For example, a program that is not performing as expected can revisit their hiring mechanism to ensure future hires can better help the team deliver results. However, asking the current team to work more efficiently (better intention) is most likely not going to help.

Here are some tips from the training that I find helpful:

1. What are the differences between the current state and my desired output?

2. What existing products/processes can I add to in order to achieve my desired output?

3. Who would be good to partner with on a brainstorming session about my idea?

At the end of the day, making changes is still hard. I hope I can continue to learn more about ways to improve mechanisms as I gain experience working in the industry (and share with you).