Robots & Pencils is seeking a seasoned AWS AI Solutions Architect to lead the design and delivery of complex, enterprise-grade generative and agentic AI systems built on Amazon Web Services. You will architect scalable, secure, and production-ready AI platforms leveraging Amazon Bedrock, Amazon Bedrock AgentCore, AWS Strands Agents, AWS AgentCore Gateway, Nova Forge, Nova 2 Sonic, and related AWS AI/ML services.
As an AWS AI Solutions Architect, you will serve as a strategic technical advisor-translating ambiguity into structured AWS-native architectures, validating designs through hands-on prototyping, and ensuring every solution aligns with the AWS Well-Architected Framework (including ML Lens) while delivering measurable business value.
Key Responsibilities Client Engagement & AWS Solutions Architecture - Serve as the primary AWS AI architecture partner for strategic clients, driving generative and agentic AI system design from discovery through production.
- Lead architecture design using Amazon Bedrock (including foundation models and custom models), Bedrock AgentCore, AWS Strands Agents, and AWS AgentCore Gateway.
- Design advanced RAG, Agentic RAG, and multi-agent orchestration architectures leveraging AWS-native services such as Lambda, Step Functions, API Gateway, DynamoDB, Aurora (pgvector), and OpenSearch.
- Produce AWS reference architectures, architecture decision records (ADRs), and implementation roadmaps aligned to business objectives.
- Validate feasibility through hands-on prototyping in Python using Bedrock SDKs, SageMaker, and serverless services.
- Ensure architectures follow AWS security best practices (IAM, KMS, VPC, PrivateLink) and cost optimization principles.
Outcome Ownership & Business Impact - Own architectural integrity from concept through production deployment on AWS.
- Align solutions with AWS Well-Architected Framework pillars: Operational Excellence, Security, Reliability, Performance Efficiency, Cost Optimization, and Sustainability.
- Guide clients through tradeoff decisions across model selection (Bedrock FMs vs custom SageMaker models), latency, cost, governance, and compliance.
Accelerate time-to-value through reusable AWS accelerators, Infrastructure as Code (CloudFormation/Terraform/CDK), and CI/CD automation. - Continuously evaluate emerging AWS AI capabilities (Nova Forge, Nova 2 Sonic, Bedrock updates, and new AgentCore capabilities).
Engineering Leadership & Delivery Excellence - Provide architectural oversight to Forward Deployed Engineers and AWS delivery teams.
- Establish best practices for MLOps on AWS including model lifecycle management, monitoring, and observability using SageMaker, CloudWatch, CloudTrail, and AWS Config.
- Define governance, responsible AI guardrails, Bedrock Guardrails configuration, and security controls for enterprise environments.
- Mentor engineers on AWS AI service integration, distributed systems design, and secure multi-account strategies.
- Make principled tradeoffs under constraints related to privacy, compliance (SOC2, HIPAA, GDPR), cost, and operational complexity.
Cross-Functional Collaboration - Partner with internal product, engineering, research, and customer success teams to evolve AWS-based AI offerings.
- Contribute AWS reference architectures and reusable infrastructure modules to internal accelerators.
- Support pre-sales engagements including architecture workshops, AWS migration strategy, and solution scoping.
- Collaborate across distributed teams and client stakeholders across North America.
Required Skills & Qualifications - Bachelor's degree in Computer Science, Engineering, or equivalent experience.
- 7-10+ years of experience in software engineering or cloud architecture with deep AWS ownership.
- Deep expertise in Amazon Bedrock, Bedrock AgentCore, AWS Strands Agents, AgentCore Gateway, and related AWS AI services.
- Strong familiarity with SageMaker (training, deployment, pipelines), deep learning fundamentals, and model fine-tuning strategies.
- Experience architecting RAG, multi-agent, and orchestration systems using AWS-native services.
- Strong knowledge of distributed systems, event-driven architectures, and serverless patterns.
- Proficiency with Infrastructure as Code (AWS CDK, CloudFormation, Terraform).
- Hands-on development capability in Python and AWS SDKs.
- Experience implementing observability and monitoring strategies in AWS environments.
- Proven success leading enterprise-scale AWS transformations.
- Exceptional communication skills for both technical and executive audiences.
- AWS Professional Certifications highly preferred (AWS Solutions Architect - Professional, AWS DevOps Engineer - Professional).
Nice to Have - AWS Specialty certifications (Machine Learning - Specialty, Security - Specialty).
- Experience with advanced agentic reasoning patterns (ReAct,CoT, Tree-of-Thoughts) implemented on Bedrock.
- Experience building secure multi-account AWS organizations using Control Tower.
- Exposure to data engineering services such as Glue, Redshift, Lake Formation.
- Consulting or professional services background.
Personal Competencies - Accountability - Owns AWS architectural direction and client outcomes with rigor.
- Adaptability - Rapidly adopts new AWS AI releases and evolving generative AI capabilities.
- Collaboration - Builds trust across engineering and executive stakeholders.
- Execution-Focused - Balances innovation with production-ready AWS delivery.
- Innovation-Minded - Experiments responsibly with emerging AWS AI services.
- Craftsmanship - Designs secure, scalable, and well-documented AWS systems.
- Leadership with Courage - Drives architectural alignment in complex environments.
- Comfort in Ambiguity - Translates unclear AI requirements into AWS-native solution architectures.
