As part of NOVA Month, I recently participated in the AWS Builders Skill Sprint, a hands-on learning challenge focused on building AI agents using the Strands SDK, Ollama, and Amazon Bedrock.
Before starting the sprint, I had heard terms like AI agents, memory systems, MCP servers, and Amazon Bedrock, but my understanding of how these pieces actually worked together was quite limited. I knew the concepts at a high level, but building them myself was a different experience altogether.
The sprint was structured in a way that gradually increased the difficulty. It started with creating a simple local AI agent using Ollama. While the first challenge seemed straightforward, it took me some time to understand how the agent, model, and system prompts interacted with each other. Instead of simply copying the code, I tried to understand why each component was necessary and how the overall flow worked.
As I moved on to the later challenges, things became more interesting and challenging. Adding tools to the agent was one of the first moments where I began to appreciate the power of agentic systems.
The memory challenge was another area where I spent extra time trying to understand the underlying concepts. Initially, I was confused about the difference between conversation history and persistent memory. After experimenting and observing the agent's behavior, I finally understood how vector-based memory allows information to be stored and retrieved even across different sessions.
The most challenging part for me was the final stages involving full-featured agents and MCP-powered systems. There were several moments when I had to revisit documentation, reread instructions, and troubleshoot issues that did not work on the first attempt. While it was frustrating at times, those moments forced me to learn more deeply rather than just rushing to complete the tasks.
One thing I particularly enjoyed about this sprint was that it was not purely theoretical. Every concept was tied to something I could build, run, and experiment with. Seeing an agent use tools, remember information, and interact with external systems made the learning process much more engaging. The sprint exposed me to concepts that I believe will become increasingly important as AI systems continue to evolve.
Overall, the AWS Builders Skill Sprint was a valuable learning experience. It challenged me, pushed me to explore unfamiliar concepts, and gave me practical exposure to technologies that I had only read about before. While I still have a lot to learn, I now feel much more confident exploring the world of AI agents and cloud-powered AI applications.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article