Building Production-Ready Agentic AI with AWS FDE: My Journey Through the Partner Pathways
In this article, I share my experience earning the AWS Partner Forward Deployed Engineer (FDE) Validated and Advanced credentials, what I learned across the pathways, and the hands-on work involved with Amazon Bedrock AgentCore, SageMaker AI, Cedar policies, observability, evaluation, memory, identity, and secure MCP tool access. I also explore how these capabilities and FDE practices can be applied to real-world use cases and production-ready agentic AI delivery.
I'm excited to share that I earned both of the new AWS Partner Forward Deployed Engineer (FDE) credentials:
✅ AWS Partner: Forward Deployed Engineer (FDE) Validated and
✅ AWS Partner: Forward Deployed Engineer (FDE) Advanced
For me, this was much more than a regular Partner training on the latest AWS AI services. The most valuable part was diving deep into the production-grade AI engineering methods and technical standards used by AWS's own FDE teams — and how our Partner engineers can apply the same AWS-defined production bar when delivering agentic AI solutions to customers.
This is especially relevant as AWS has announced a $1 billion investment in a Forward Deployed Engineering unit and extending the FDE model to AWS Partners. I see these new credentials as a great enabler for Partner and Partner engineers to validate real delivery skills and create new business revenue opportunities around production agentic AI.
Through these two credentials, you'll learn how to build production-ready agentic AI solutions for real-world use cases using nearly all AWS AI services such as Amazon Bedrock AgentCore, Amazon Bedrock, and Amazon SageMaker AI, etc.
Some examples I worked through included:
→ Building and deploying observable and secure agents with AgentCore Runtime and Observability
→ Using AgentCore Identity for fine-grained agent access control
→ Configuring AgentCore Memory for conversation persistence
→ Using Cedar policies to govern access to MCP tools through AgentCore Gateway
→ Evaluating AI customer support response performance with AgentCore Evaluation
→ Building retail recommendation agents and healthcare Prior Authorization A2A workflows
→ Deploying and fine-tuning models in SageMaker AI for a Municipal AI Innovation Hub
......
→ Using AgentCore Identity for fine-grained agent access control
→ Configuring AgentCore Memory for conversation persistence
→ Using Cedar policies to govern access to MCP tools through AgentCore Gateway
→ Evaluating AI customer support response performance with AgentCore Evaluation
→ Building retail recommendation agents and healthcare Prior Authorization A2A workflows
→ Deploying and fine-tuning models in SageMaker AI for a Municipal AI Innovation Hub
......
These skills can help you deliver AI solutions for different business use cases, both for customers and within your own organization.
Start the new AWS Partner FDE training here
The requirements to earn the credentials:
FDE Validated
• AWS Partner Network membership
• Active AWS Associate- or Professional-level certification
• 600 points (My route included 1x assessment, 2x SimuLearn labs, and 5x Microcredentials, and you'll finish it in roughly 6 hours or less.)
• Complete all required activities in at least one pathway. (There are three pathways: Ground, Orchestrate and Prove. I found Ground the most straightforward path)
• AWS Partner Network membership
• Active AWS Associate- or Professional-level certification
• 600 points (My route included 1x assessment, 2x SimuLearn labs, and 5x Microcredentials, and you'll finish it in roughly 6 hours or less.)
• Complete all required activities in at least one pathway. (There are three pathways: Ground, Orchestrate and Prove. I found Ground the most straightforward path)
FDE Advanced
• AWS Partner Network membership
• AWS Certified Generative AI Developer – Professional
• 1,000 points (1x assessment, 5x Microcredentials, and 7x SimuLearn labs)
• Complete Ground, Orchestrate, and Prove (I completed the remaining work in roughly another 12 hours and you may need less.)
• AWS Partner Network membership
• AWS Certified Generative AI Developer – Professional
• 1,000 points (1x assessment, 5x Microcredentials, and 7x SimuLearn labs)
• Complete Ground, Orchestrate, and Prove (I completed the remaining work in roughly another 12 hours and you may need less.)
Thanks to @Julia Mazuryka and @Erik Farr for sharing these valuable new credentials with the AWS Partner community.
I'd also love to share this with other AWS community developers. If you're an AWS Partner developer interested in these FDE pathways, feel free to reach out. Happy to answer questions, exchange experience, and learn from others in the community as well.
Enjoyed reading this content? Let the author know!
Your likes, comments, shares, and saves help creators reach more builders.
Loading recommendations
Loading article