Career Roadmap
AWS Forward Deployed Engineer: Zero to Hero
This roadmap guides you from AWS engineering foundations to professional Forward Deployed Engineering in the AWS ecosystem. You will build customer-facing agentic AI delivery skills, master the AWS agent stack (Amazon Bedrock AgentCore, Strands Agents, SageMaker AI), and earn the credentials that validate the FDE skill blend. SAA-C03 satisfies the FDE Validated certification prerequisite. AIF-C01 builds AI literacy. AIP-C01 satisfies the FDE Advanced prerequisite. The AWS Partner FDE pathways validate hands-on delivery in AWS-provisioned environments, not multiple-choice recall. Mid-level and senior engineers can compress the foundation steps. Use ExamOS practice quizzes to track progress at every stage.
Who is this roadmap for?
This roadmap is designed for AWS Engineers and Cloud Developers transitioning into customer-facing delivery roles, as well as Solutions Architects who want to move from designing to shipping. Additionally, it is tailored for AI Engineers who want to deploy agentic systems inside enterprise AWS environments, as well as Senior Engineers who want a faster track into AWS FDE roles without restarting their careers.
Skills You'll Develop
Write production Python and deploy to AWS using containers, Lambda, and infrastructure as code.
Build and operate agents with Amazon Bedrock AgentCore, Strands Agents, and SageMaker AI.
Connect agents to authoritative business data, domain ontologies, and enterprise context.
Instrument agents with AgentCore Observability, evaluation, Cedar policies, and human-in-the-loop controls.
Turn ambiguous business problems into written problem statements and scoped deliverables.
Run discovery sessions, deliver demos, write executive briefs, and drive adoption.
Target Roles in this Roadmap
- Forward Deployed Engineer (AWS): Embeds with customers to build and ship production agentic AI systems on AWS
- Forward Deployed AI Engineer: Specializes in Bedrock AgentCore, RAG, and multi-agent workflows inside enterprise AWS accounts
- Senior Forward Deployed Engineer: Owns the technical success of multi-customer agentic AI engagements and mentors junior FDEs
- AI Solutions Architect (FDE-adjacent): Designs systems and hands off to delivery teams; FDEs often move into or out of this role
- Product Manager (FDE exit path): FDEs who move into product roles bring direct customer context and deployment experience
Typical Employer Categories
- AWS Partners and Consulting Firms: Deloitte, Accenture, EPAM, and APN partners running Partner-Led FDE practices
- AWS and Its FDE Organization: The $1 billion Forward Deployed Engineering organization AWS announced in June 2026
- Enterprise Software and SaaS Companies: Organizations building AWS-based agentic AI products that require customer-facing deployment
- Financial Services and Healthcare: Regulated enterprises deploying agents inside AWS compliance boundaries
- Managed Service Providers and System Integrators: Partners delivering production agentic AI under the FDE motion rather than traditional project delivery
The Certification Path
Recommended Path
| Cert | When | Why |
|---|---|---|
| AWS Certified Solutions Architect Associate (SAA-C03) | Month 1-3 | Satisfies the active AWS Associate or Professional certification prerequisite for FDE Validated. Builds the architectural trade-off reasoning FDE engagements demand. |
| AWS Certified AI Practitioner (AIF-C01) | Month 1-2 | Foundational AI and generative AI literacy. Optional but useful if you are new to AI workloads on AWS. The exam code is confirmed as AIF-C01 as of 2026. |
| AWS Certified Generative AI Developer – Professional (AIP-C01) | Month 4-7 | Required prerequisite for FDE Advanced. Validates foundation model integration, RAG, agentic implementation, safety, and operational efficiency. Passing score is 750/1,000. This is a professional-level exam. |
| AWS Partner FDE Validated | Month 5-8 | The entry FDE credential. Requires APN membership, an active AWS Associate or Professional certification, completion of all required validation activities in at least one pathway, and 600 points across Ground, Orchestrate, and Prove. |
| AWS Partner FDE Advanced | Month 8-12 | The senior FDE credential. Requires APN membership, AIP-C01, and completion of the published validation requirements across all three pathways plus 1,000 points. |
Optional Foundation
| Cert | When | Why |
|---|---|---|
| AWS Certified Cloud Practitioner (CLF-C02) | Month 1 | Optional warm-up. Only needed if you are completely new to AWS. |
Specialization Path
| Cert | When | Why |
|---|---|---|
| AWS Certified Machine Learning Engineer – Associate (MLA-C02) | Month 9-12 | ML engineering depth for FDEs whose customers run ML and LLMOps workloads. MLA-C01 retired in September 2026. MLA-C02 beta opened September 2026, with the standard version expected in early 2027. Verify availability in your region before booking. |
| AWS Certified Solutions Architect – Professional (SAP-C02) | Month 9-12 | Enterprise architecture depth for senior FDEs designing multi-account, multi-region customer environments. |
| AWS Certified DevOps Engineer – Professional (DOP-C02) | Month 9-12 | Deployment automation depth for platform-focused FDE roles. |
| HashiCorp Terraform Associate | Month 3-5 | Infrastructure as code is standard in FDE deployments, even in AWS-native shops. |
| CKA (Certified Kubernetes Administrator) | Month 5-7 | Container orchestration depth for customers running agents and models on EKS. |
Milestones: Junior → Mid → Senior
| Level | Milestone | When |
|---|---|---|
| Entry Level | SAA-C03 + production AWS engineering fluency + basic agent deployment understanding | Month 3-4 |
| Practitioner Level | AIP-C01 + FDE Validated + hands-on customer-facing agentic AI deployment project | Month 7-8 |
| Specialist Level | FDE Advanced + MLA-C02 or SAP-C02 + multi-engagement delivery experience | Month 10-12 |
Step 1 - AWS engineering and AI foundations
Build the AWS engineering fundamentals and AI literacy that every FDE engagement depends on. SAA-C03 is the fastest route to satisfying the active AWS certification prerequisite for FDE Validated. If you already hold an Associate or Professional certification, you can skip the exam and spend this time on hands-on agent work.
~1-3 months~1-3 months
Step 1 - AWS engineering and AI foundations
Build the AWS engineering fundamentals and AI literacy that every FDE engagement depends on. SAA-C03 is the fastest route to satisfying the active AWS certification prerequisite for FDE Validated. If you already hold an Associate or Professional certification, you can skip the exam and spend this time on hands-on agent work.
- AWS core services: VPCs, subnets, security groups, IAM, EC2, S3, Lambda, and RDS
- Architectural reasoning: secure, resilient, high-performing, and cost-optimized designs
- Containers and compute: ECS, EKS basics, Lambda, and event-driven patterns
- Infrastructure as code: CloudFormation and Terraform basics for repeatable deployments
- Production Python: APIs, error handling, boto3, and working inside unfamiliar customer codebases
- AI and generative AI literacy: foundation models, tokens, embeddings, prompt design, and responsible AI
- AWS AI service landscape: Bedrock, SageMaker AI, and where each fits
Certifications
💡 FDE work starts with production engineering inside customer environments. If you cannot ship and debug AWS workloads, customer-facing skills will not compensate.
💡 FDE Validated accepts an active Associate or Professional AWS certification. SAA-C03 is the most common and most broadly useful choice for FDE roles.
💡 AIF-C01 is optional if you already work with AI workloads. The exam code has changed before; verify the current version on the AWS certification page before booking.
🏁 Entry Level Checkpoint: You can deploy a production-grade AWS application with IAM, networking, and monitoring, and you can explain the AWS AI service landscape.
🛠 Project Ideas
- ▸Deploy a containerized API to ECS Fargate with IAM task roles and access to S3 and RDS.
- ▸Write a Terraform or CloudFormation template that provisions a VPC, Lambda function, and DynamoDB table. Deploy it to a test account.
- ▸Call Amazon Bedrock from a Python script using boto3 and IAM credentials instead of a static access key.
Step 2 - Agentic AI engineering on AWS
Build the applied agentic AI engineering skills that the AWS FDE pathways actually validate. This is not model training. It is building agents that connect to authoritative enterprise data, coordinate workflows, recover from failures, and meet enterprise security requirements inside AWS-provisioned environments.
~2-3 months~2-3 months
Step 2 - Agentic AI engineering on AWS
Build the applied agentic AI engineering skills that the AWS FDE pathways actually validate. This is not model training. It is building agents that connect to authoritative enterprise data, coordinate workflows, recover from failures, and meet enterprise security requirements inside AWS-provisioned environments.
- Amazon Bedrock: foundation model access, model selection, inference APIs, and Guardrails
- RAG with Bedrock Knowledge Bases: ingestion, chunking, embeddings, vector stores, and retrieval evaluation
- Amazon Bedrock AgentCore: Runtime, Identity, Memory, Gateway, Observability, and Evaluation
- Agent frameworks: Strands Agents, AgentCore CLI, and code-based versus config-based agents
- Tool access and governance: MCP servers, AgentCore Gateway, and Cedar policies for tool authorization
- Multi-agent patterns: Agent2Agent (A2A) workflows, orchestration, and handoff
- Model deployment and tuning: SageMaker AI endpoints, inference options, and fine-tuning
- Observability: OpenTelemetry, CloudWatch, and tracing across agent steps and tool calls
💡 The AWS FDE pathways are built around exactly this stack. Candidate reports from the FDE pathways mention working with AgentCore Runtime, AgentCore Identity, AgentCore Memory, Cedar policies, AgentCore Gateway, AgentCore Evaluation, SageMaker AI, and A2A workflows.
💡 Build hands-on familiarity before attempting any pathway validation. The credentials are earned through hands-on validation in AWS-provisioned environments, not solely through course completion.
🏁 Practitioner Level Checkpoint 1: You can build, deploy, instrument, and evaluate an agent on Amazon Bedrock AgentCore with grounded data and governed tool access.
🛠 Project Ideas
- ▸Build a RAG agent in AgentCore that answers questions over a document set and cite retrieved sources. Instrument it with AgentCore Observability.
- ▸Build an agent with three tools behind AgentCore Gateway and restrict tool access with Cedar policies. Test what happens when a call is denied.
- ▸Configure AgentCore Memory so a multi-turn conversation retains context, then evaluate response quality with AgentCore Evaluation.
Step 3 - FDE Validated, AIP-C01, and customer-facing delivery
Earn AIP-C01, complete one FDE pathway, and build the customer-facing delivery skills that separate FDEs from backend AI engineers. FDE Validated requires APN membership, an active Associate or Professional certification, all required validation activities in at least one pathway, and 600 points across Ground, Orchestrate, and Prove.
~2-3 months~2-3 months
Step 3 - FDE Validated, AIP-C01, and customer-facing delivery
Earn AIP-C01, complete one FDE pathway, and build the customer-facing delivery skills that separate FDEs from backend AI engineers. FDE Validated requires APN membership, an active Associate or Professional certification, all required validation activities in at least one pathway, and 600 points across Ground, Orchestrate, and Prove.
- AIP-C01 domains: foundation model integration, data management, and compliance (31%); implementation and integration (26%); AI safety, security, and governance (20%); operational efficiency and optimization (12%); testing, validation, and troubleshooting (11%)
- Ground pathway: trusted enterprise data, systems, domain ontologies, and business context
- Orchestrate pathway: dependable agents, tools, workflows, identity, and recovery
- Prove pathway: evaluation, observability, security, policy, and human control
- Customer discovery: turning vague pain points into a written problem statement the customer agrees with
- Rapid prototyping: shipping a useful proof of concept in days, not months
- Enterprise integration: legacy systems, SSO, API gateways, and event-driven patterns
- Stakeholder communication: demos, executive updates, written briefs, and managing adoption
Certifications
💡 FDE Validated requires APN membership, an active Associate or Professional AWS certification, completion of all required validation activities in at least one pathway, and 600 points. One candidate reported finishing in roughly 6 hours of validation activity once the prerequisites and skills were in place.
💡 AIP-C01 is not required for FDE Validated. It is required for FDE Advanced, so candidates targeting the senior credential should start it here rather than at the end. The AIP-C01 passing score is 750.
💡 The FDE role is roughly 60% customer-facing, 30% building deployment-specific code, and 10% internal work. Technical depth gets you into the room; communication keeps you effective.
🏁 Practitioner Level Checkpoint 2: You hold FDE Validated and AIP-C01, and you can run a discovery session, write a scoped brief, and deliver a working agentic AI proof of concept inside a customer's AWS environment.
🛠 Project Ideas
- ▸Complete the Ground pathway first. Candidate reports describe Ground as the most straightforward entry point, and it maps directly to the enterprise data work most FDE engagements start with.
- ▸Build a proof of concept in a simulated customer environment with a legacy database, an SSO constraint, and a two-week timeline.
- ▸Write a discovery brief and a scoped delivery plan for a real or simulated customer problem.
Step 4 - FDE Advanced and specialization
Move from delivery on individual engagements to leading multi-customer agentic AI deployments. FDE Advanced requires APN membership, AIP-C01, and completion of the published validation requirements across all three pathways plus 1,000 points. Choose the specialization that matches the FDE roles you are targeting.
~3-6 months (ongoing)~3-6 months (ongoing)
Step 4 - FDE Advanced and specialization
Move from delivery on individual engagements to leading multi-customer agentic AI deployments. FDE Advanced requires APN membership, AIP-C01, and completion of the published validation requirements across all three pathways plus 1,000 points. Choose the specialization that matches the FDE roles you are targeting.
- **FDE Advanced requirements**: APN membership, AIP-C01, all three pathways (Ground, Orchestrate, Prove), and 1,000 points across assessments, microcredentials, and SimuLearn labs
- Production agentic AI at scale: agent identity, policy enforcement, recovery, and multi-agent orchestration across customer environments
- Enterprise architecture: multi-account governance, Control Tower, and hybrid connectivity for AI workloads
- ML and LLMOps depth: SageMaker AI pipelines, model monitoring, and drift detection for customers running their own models
- Multi-customer engagement management: standardizing delivery, building reusable connectors, and packaging accelerators
- Coaching and mentoring: reviewing implementations, setting technical standards, and growing the FDE practice
- **Career exits**: solutions architect, product management, or engineering leadership
Certifications
💡 Senior FDEs are judged on outcomes, not code shipped. FDE Advanced is the credential that signals you can operate across the full production agentic AI surface, not just one pathway.
💡 MLA-C02 replaced MLA-C01 in September 2026. The beta opened in September 2026, and the standard version is expected in early 2027. Verify exam availability in your region before booking.
💡 SAP-C02 adds architecture depth for senior FDEs designing enterprise customer environments. DOP-C02 adds deployment automation depth for platform-focused FDE roles.
🏁 Specialist Level Checkpoint: You hold FDE Advanced, can lead multi-customer agentic AI engagements, mentor junior FDEs, and influence the delivery practice.
🛠 Project Ideas
- ▸Complete the remaining two pathways. Candidate reports describe roughly another 12 hours of validation work to finish all three after starting with one.
- ▸Lead a multi-week customer deployment end to end: discovery, architecture, build, hardening, and handover.
- ▸Build a reusable accelerator (an AgentCore agent template, a Cedar policy pack, or a Terraform landing zone module) that other FDEs can use.
Final Step - What this path actually builds
This roadmap builds an AWS engineer who can ship production code inside a customer's account, build and operate agents on Amazon Bedrock AgentCore, and prove that those agents are reliable, observable, and governable. SAA-C03 satisfies the FDE Validated prerequisite. AIP-C01 satisfies the FDE Advanced prerequisite. The AWS Partner FDE pathways validate hands-on delivery in AWS-provisioned environments, not multiple-choice recall, which is why they carry more signal than a course certificate. FDE Validated proves depth in at least one pathway. FDE Advanced proves depth across Ground, Orchestrate, and Prove. The role is not a renamed solutions architect role. Solutions architects design and hand off. FDEs write code, stay until production works, and are judged on measurable business outcomes. Mid-level and senior engineers can compress the foundation steps. The technical bar is real, but the career differentiator is customer-facing delivery judgment.
Final Step - What this path actually builds
This roadmap builds an AWS engineer who can ship production code inside a customer's account, build and operate agents on Amazon Bedrock AgentCore, and prove that those agents are reliable, observable, and governable. SAA-C03 satisfies the FDE Validated prerequisite. AIP-C01 satisfies the FDE Advanced prerequisite. The AWS Partner FDE pathways validate hands-on delivery in AWS-provisioned environments, not multiple-choice recall, which is why they carry more signal than a course certificate. FDE Validated proves depth in at least one pathway. FDE Advanced proves depth across Ground, Orchestrate, and Prove. The role is not a renamed solutions architect role. Solutions architects design and hand off. FDEs write code, stay until production works, and are judged on measurable business outcomes. Mid-level and senior engineers can compress the foundation steps. The technical bar is real, but the career differentiator is customer-facing delivery judgment.
Certifications
Final Thoughts
💡 Total: 6-10 months for junior engineers; 3-5 months for mid-level engineers with AWS experience; 2-3 months for senior engineers who already lead AWS deployments
💡 AWS foundations: 2-3 months for junior; 4-6 weeks for mid-level; 2 weeks for senior
💡 Agentic AI engineering: 2-3 months for all levels; senior engineers can compress this if they already deploy agents on Bedrock
💡 FDE Validated: roughly 6 hours of validation activity once prerequisites and skills are in place, plus the AIP-C01 study time if you are targeting FDE Advanced
💡 FDE Advanced: roughly 12 additional hours of validation activity to complete all three pathways after starting with one
💡 Readiness: Consistent 80%+ on Legend mode across five sessions is your signal to book any certification exam
Honest Timeline
| Path | Minimum Study Pace | More Realistic Pace |
|---|---|---|
| Junior AWS engineer (0-2 years) | ~6 months | ~9-10 months |
| Mid-level AWS engineer (2-5 years) | ~3 months | ~5 months |
| Senior engineer (5+ years, production AWS experience) | ~1 month | ~2-3 months |
| Senior engineer targeting FDE Advanced plus MLA-C02 or SAP-C02 | ~4 months | ~6 months |