← back
Introducing Strands Agents, an Open Source AI Agents SDK — Suman Debnath, AWS
Takeaway
Strands treats agent building as just-model-plus-tools, leveraging MCP and LLM reasoning rather than hand-written scaffolding.
Summary
- AWS's open-source Strands SDK reduces an agent to two strands: a model and tools — no system-prompt scaffolding required, betting on model intelligence to do the reasoning.
- Defaults to Bedrock (Claude 3.7) but plugs into LiteLLM, Ollama, Langfuse and any third-party provider; ships with built-in tools for file read/write, speech, etc.
- Demo 1: agent reads a file, summarizes it, writes the summary to disk and speaks the output — all via default tools with no custom code.
- Demo 2: connects via MCP client to a Manim server to generate 3Blue1Brown-style math visualizations from natural-language prompts; custom Python functions become tools by decorating with @tool.
strandsagents-sdkmcp
Original description
Building AI agents used to require complex orchestration, extensive scaffolding, and months of tuning. With Strands Agents, an open source SDK from AWS. You can now build, test, and deploy intelligent agents in just a few lines of code. This session introduces the model-driven approach behind Strands, where a model, a prompt, and a set of tools are all you need to create powerful, production-ready agents. Learn how Strands leverages modern foundation models to handle reasoning, tool use, and reflection, reducing development time from months to days. About Suman Debnath Suman Debnath is a Principal Machine Learning Advocate at Amazon Web Services. Currently, his focus is on Supervised Learning, Natural Language Processing (NLP), Large Language Models (LLMs), and Retrieval Augmented Generation (RAG). Suman is committed to leveraging open-source tools like LangChain, PyTorch, Numpy, and Pandas for advancing machine learning. He has developed performance benchmarking and monitoring tools for distributed storage systems. Suman has spoken at over 100 global events, including AWS re:Invent, AI Engineer Summit, PyCon, PyData, ODSC, and meetups across multiple countries. Recorded at the AI Engineer World's Fair in San Francisco. Stay up to date on our upcoming events and content by joining our newsletter here: https://www.ai.engineer/newsletter