AI Solutions Architect

Kelly Science, Engineering, Technology & Telecom; New York (NY); On-site; USD 180,000 - 230,000; Full time

Job summary

Kelly Science, Engineering, Technology & Telecom is seeking an experienced engineer to architect and build multi-step agentic workflows using AWS Agent Core and related tooling. You will design secure, high-throughput API layers, manage the underlying AWS infra, and implement guardrails, observability, and zero-trust security to enable enterprise AI deployment.

Responsibilities include integrating LLMs via Bedrock, orchestrating Lambda/Step Functions, and establishing reusable reference

Qualifications

• 5+ years of professional software/solutions engineering experience, including 3+ years of production-grade AWS architecture.
• Hands-on experience building complex, multi-agent workflows or tool-using autonomous systems.
• Practical experience with AWS Agent Core or related orchestration frameworks (LangGraph, AutoGen, Bedrock Agents).
• Advanced API management with AWS API Gateway, including REST/HTTP API configuration, custom authorizers, throttling, and transformations.

Responsibilities

• Agentic Workflow Development: architect and build multi-step agentic workflows using AWS Agent Core or equivalents, with advanced tool use and memory retrieval.
• API & Connectivity Infrastructure: design secure, high-throughput API layers using AWS API Gateway and MCP for internal and external exposure.
• Core Cloud Engineering: manage AWS infrastructure supporting agents, optimizing Lambda/ECS/EKS, Step Functions, EventBridge, storage (S3, DynamoDB, RDS/Aurora, Snowflake).
• LLM Integration & Prompt Engineering: integrate foundation models via Bedrock and other endpoints, mastering function calling and context windows.
• Guardrails & Observability: implement enterprise controls for reliability, safety, cost control, including AWS X-Ray, toxic content filtering, HITL checkpoints.
• Zero-Trust Security: design robust authentication/authorization (IAM, OAuth, Lambda Authorizers) for agent interfaces to services and APIs.

Skills

Agentic AI

AWS Architecture

Framework Proficiency

API Management

Security & IAM

DevOps & IaC

Programming: Python/TS

Tools

AWS

Terraform

AWS CDK

CloudFormation

Lambda

ECS/EKS

Job description

• Agentic Workflow Development: Architect and build multi-step agentic workflows using AWS Agent Core (or equivalents), integrating advanced tool use, memory preservation, retrieval, and multi-step reasoning capabilities.
• API & Connectivity Infrastructure: Design and implement secure, high-throughput API layers using AWS API Gateway and the Model Context Protocol (MCP) to safely expose agent capabilities to internal systems and external consumers.
• Core Cloud Engineering: Build and manage the underlying AWS infrastructure supporting your agents, optimizing compute (Lambda, ECS, EKS), event-driven orchestration (Step Functions, EventBridge), storage (S3, DynamoDB, RDS/Aurora, Snowflake), and messaging (SQS, SNS, Kafka).
• LLM Integration & Prompt Engineering: Integrate foundation models via Amazon Bedrock (and other endpoints), mastering structured function calling, prompt/tool boundaries, and context window optimization.
• Guardrails & Observability: Implement enterprise-grade controls for reliability, safety, and cost control—including distributed tracing (AWS X-Ray), evaluation frameworks, toxic content filtering, and Human-in-the-Loop (HITL) validation checkpoints.
• Zero-Trust Security: Design rigid authentication and authorization patterns (IAM, OAuth, Lambda Authorizers) governing how autonomous agents interface with internal microservices and third-party APIs.
• Organizational Impact: Establish reusable reference architectures, design patterns, and deployment best practices to accelerate agentic AI development across the entire enterprise.

Required Qualifications

• Experience: 5+ years of professional software/solutions engineering experience, including 3+ years of production-grade AWS architecture and development.
• Agentic AI Expertise: Demonstrated, hands-on experience building complex, multi-agent workflows or tool-using autonomous systems (beyond basic prompt engineering or short-lived POCs).
• Framework Proficiency: Practical experience with AWS Agent Core, or a proven ability to scale quickly into it using related orchestration frameworks (e.g., LangGraph, AutoGen, Bedrock Agents).
• Advanced API Management: Deep expertise with AWS API Gateway, including REST/HTTP API configuration, custom authorizers, throttling policies, and request/response transformations.
• AWS Systems Mastery: Strong fluency with the core AWS ecosystem, specifically Lambda, Step Functions, DynamoDB, S3, EventBridge, IAM, and CloudWatch.
• Modern Engineering Practices: Strong software engineering fundamentals, including API contract design, automated testing, CI/CD pipelines, and Infrastructure as Code (IaC) using Terraform, AWS CDK, or CloudFormation.
• Languages: Ability to write clean, maintainable, production-quality code in Python and/or TypeScript (Node.js).