Building for the Future

Hey Builders,

This month marks my 3-year anniversary at AWS, and I'm incredibly excited about what we've accomplished and where we're headed together! To celebrate, I want to talk about "Building for the Future."

As we evolve in our roles, we're seeing an interesting bifurcation within our community—those who are leaning into AI and those who aren't. And that's okay. Leveraging AI is entirely up to you, the builder. But for those who want to build with AI, who want to use it to improve, that's where my focus lies.

AI presents a tremendous opportunity, but there's no predefined playbook for integrating it into your organization. As Ethan Mollick put it:

Nobody has special information about how to best use AI at your company, or a playbook for how to integrate it into your organization.

We’re all discovering this together, and it’s up to each of us to find out how to make these tools work best in our own contexts.

For those who are curious about AI, who want to harness it to make better tools, create more efficiently, and scale ideas, the future is full of possibilities.

We’re building a future where personalized AI assistants can support us in developing code, crafting content, and solving complex problems. It's a future where AI can be our creative collaborator and our partner in problem-solving.

Building for the Future: Systems, Jobs, and Tasks

To effectively build for the future, it's important to break down our work into systems, jobs, and tasks. Understanding these layers helps us see where AI can have the most impact and how we can evolve our roles to focus on what matters most.

A system is the overarching structure or environment in which work takes place. Within a system, there are specific jobs that contribute to the overall goals of that system. Each job, in turn, is made up of various tasks, which are the individual actions or responsibilities that make up the day-to-day work. By understanding how systems, jobs, and tasks interrelate, we can see where AI can best be leveraged to improve efficiency, productivity, and innovation at each level.

When working with AI, I often ask myself these guiding questions:

1. How do I invite AI to the table? This helps me determine where AI can add value, by identifying tasks that AI can assist with or augment, ensuring that AI is leveraged to increase efficiency and creativity.
2. How do I act as the human in the loop? This question helps me maintain control and oversight over AI's contributions. It’s about ensuring that I provide context, make critical decisions, and verify AI outputs to maintain high quality.
3. How do I treat AI as a human? I approach AI like a collaborator by giving it a persona and treating it as a partner. This makes interactions more natural and helps me use AI in a way that complements my work.

One of the key ideas that helps bridge the concepts of systems, jobs, and tasks to the broader discussion of our evolving roles is how we think about value. As our work evolves with AI integration, we start to see a shift in the kind of value we provide.

As Naval Ravikant once said:

I would love to be paid purely for my judgment, not for any work. I want a robot, capital, or computer to do the work, but I want to be paid for my judgment.

This sentiment captures the essence of what building for the future means: letting AI handle the repetitive tasks while we focus on the strategic decisions, creative thinking, and human connections that truly add value.

Building for the Future: Developer Advocate

To build for the future, it's important to understand how AI can integrate into our daily work at multiple levels. Let me share how I approach my role as a Developer Advocate through this lens.

System: Developer Experience

The Developer Experience system is about winning the hearts and minds of builders. It involves everything from creating tools and resources that developers need, to providing support and guidance as they work. The goal is to create an environment where developers can thrive, and AI is becoming an increasingly critical part of this system.

Job: Developer Advocate

As a Developer Advocate, my job is to educate and inspire developers at scale. This means creating technical content, providing thought leadership, and building a community around our tools and platforms. My focus is on empowering developers by sharing knowledge and fostering connections. In the future, AI will help us do this more efficiently, but the core of the job—building relationships and providing guidance—will remain human-driven.

Tasks

To truly understand how AI can fit into our daily responsibilities, it's helpful to break down the specific tasks that make up my role as a Developer Advocate. Below, I'll explore how AI can support each task, the limitations we face, and where the human element is irreplaceable.

1. Writing Code

• Task Type: Cyborg (AI as an augmentation tool)
• How do I invite AI to the table? I use a chat interface or in-line code suggestions to help write code more efficiently. Tools like Amazon Q and Cursor are great here.
• How do I act as the human in the loop? I review the code, provide context-specific suggestions, and ensure that the AI's output fits within the broader system in which it operates.
• How do I treat AI as a human? I provide relevant documentation, and explain my preferences. I show it past examples of code that works for me.
• Current Limitations: The AI can sometimes provide incorrect code and lacks knowledge of the system in which the code is deployed. This is why being a developer still matters.

2. Creating Written Content

• Task Type: Cyborg (AI as an augmentation tool)
• How do I invite AI to the table? I use AI tools to brainstorm, create outlines, and assist with drafting content.
• How do I act as the human in the loop? I refine the content generated by AI to align with my audience's needs and ensure that it captures the right tone.
• How do I treat AI as a human? I interact with it as a collaborator, providing feedback to improve future suggestions. I show my previous work and rules for my writing.
• Current Limitations: AI-generated content can sometimes lack depth or miss the nuance of a particular topic.

(my last blog post was about how I used AI to help update an open source rep in 30 minutes)

3. Creating Video Content

• Task Type: Centaur (Human-AI collaboration)
• How do I invite AI to the table? AI helps with generating scripts and suggesting visuals.
• How do I act as the human in the loop? I adjust the AI’s suggestions to ensure they align with the style and message I want to convey.
• Current Limitations: AI cannot fully capture my personal style and tone, which makes human intervention essential.

(Are you following awsdevlopers on Instagram?)

4. Connecting with the Community

• Task Type: Just Me (No AI involvement)
• Description: Genuine, real-time interactions with the community, such as attending events, answering questions, and building relationships, are areas where I am fully involved without AI assistance.

Potential Role Changes with AI Assistance

As AI becomes more integrated, my role will focus increasingly on strategic planning, thought leadership, and community building. AI will handle more of the operational aspects of content creation and code development, allowing me to prioritize high-level decision-making, ethical considerations, and developer relationships. The goal is to create a symbiotic relationship between human expertise and AI capabilities, building a future where both can thrive.

Potential System Changes with AI Assistance

The Developer Experience system itself will evolve into something more personalized and efficient, where AI tools adapt to individual developers’ learning styles and preferences. This will enable faster feedback loops and more intuitive support systems, enhancing how we build, learn, and collaborate. The focus will remain on winning the hearts and minds of builders, with AI acting as an enabler rather than a replacement.

As we move forward, I am more committed than ever to engaging with the community, sharing insights, and learning together. This journey is about building relationships that inspire innovation and foster collaboration.

Community Events

If you want to continue the conversation or catch up in person, I’ll be at some upcoming community events:

• AWS San Francisco Loft Gen AI Day - Oct 15

I'll be giving a workshop on building AI Agents.

• AWS DMV Community Day - Oct 18

I'll be giving a Game Day workshop where you get to build your own AI Fighter and compete against others!

• DC Startup Week - Oct 21–Oct 25

Will have a workshop and panel discussion!

• AWS re:Invent - Dec 2–Dec 6

Lots of fun here, I have 3 sessions (reserve your seats now) and a GameDay.

Hope to see you soon.

Until then, keep building!