← Back to jobs
New York, NY, USA
No related jobs found
Agent Engineering Skill Set: Experience with framework orchestration (e.g., LangGraph, CrewAI, or visual workflow builders) to construct autonomous pipelines and configure tool contract designs for reliable API executions. Proven expertise in retrieval engineering (RAG) for precise context injection, paired with the ability to implement robust observability and security guardrails to ensure system reliability and safety.
Workflow Design Experience: Hands-on experience in AI workflow design and implementation, including leading discovery workshops with business and technical stakeholders to map current-state processes and identify automation opportunities. Proven ability to test, troubleshoot, and refine AI agents based on business requirements and user feedback to ensure seamless adoption.
Domain & Business Knowledge: Strong business process analysis skills specifically tailored to accounting and the month-end/year-end book close process are preferred.
Location: Based in the San Francisco Bay Area or the New York tri-state area. Candidates must be willing to co-locate at Intuit's Mountain View or New York office respectively, and to travel to Intuit's customers.
AWS Forward Deployed Engineer
Agentic AI and Generative AI
Role Purpose
The AWS Forward Deployed Engineer is a client-facing technical leader who discovers, designs, builds, and industrializes enterprise Generative AI and agentic AI solutions on AWS.
The role combines client consulting, business-process expertise, hands-on AI engineering, AWS architecture, security, and production ownership. The FDE works with business stakeholders, product teams, architects, data scientists, security teams, and engineers to convert priority business problems into secure, scalable, observable, and measurable AI solutions.
The FDE owns the journey from business discovery and rapid prototyping through production deployment, adoption, and continuous improvement.
Key Responsibilities
1. Client Discovery and Business Transformation
Lead discovery and process-mapping workshops with business, operations, product, data, risk, security, and technology stakeholders.
Understand business objectives, workflows, pain points, controls, data availability, and expected outcomes.
Identify high-value opportunities for AI augmentation, agent-assisted work, and workflow automation.
Translate business problems into use cases, user journeys, technical scope, acceptance criteria, dependencies, and measurable outcomes.
Build rapid AWS-based prototypes to validate feasibility and demonstrate business value.
Integrate solutions into existing business processes and enterprise applications to drive adoption.
2. Domain and Process Expertise
Develop a working understanding of the client’s industry, operating processes, terminology, policies, controls, and regulatory requirements.
Collaborate with domain experts to translate business knowledge into prompts, retrieval strategies, tools, rules, and validation controls.
Design agent workflows that address exceptions, approvals, segregation of duties, and human intervention.
Prioritize opportunities based on business value, technical feasibility, data readiness, operational risk, and adoption potential.
Define outcome measures such as processing-time reduction, manual-effort reduction, workflow completion, accuracy, user adoption, and cost per transaction.
3. AWS GenAI and Agentic Architecture
Design enterprise GenAI and agentic solutions using Amazon Bedrock, Amazon Bedrock AgentCore, and AWS cloud-native services.
Architect conversational AI, document intelligence, knowledge search, structured-output, tool-calling, and workflow-automation solutions.
Select models based on accuracy, latency, context requirements, security, and cost.
Design controlled agent workflows covering planning, memory, tool execution, validation, retries, exceptions, approvals, escalation, and fallback.
Balance agent autonomy with enterprise requirements for predictability, security, auditability, and human oversight.
Use multi-agent architectures only when they offer measurable advantages over deterministic or single-agent designs.
img-5fc7a38a-bfd1-48fb-82bc-ca41030e732c.png
Engineering, Industrialization, and Governance
AI, Retrieval, and Application Engineering
Design grounded Retrieval-Augmented Generation pipelines for structured and unstructured enterprise information.
Implement ingestion, chunking, embeddings, hybrid retrieval, reranking, metadata filtering, entitlement controls, and citations.
Use Amazon Bedrock Knowledge Bases, Amazon OpenSearch Service, Amazon Aurora PostgreSQL vector capabilities, Amazon S3, and other approved AWS services.
Build production-quality AI applications using Python, Java, FastAPI, Spring Boot, LangGraph, LangChain, CrewAI, LlamaIndex, Spring AI, or LangChain4j.
Develop reusable prompts, tools, connectors, evaluators, agent components, and workflow templates.
Integrate AI solutions with APIs, databases, microservices, messaging platforms, enterprise applications, and legacy systems.
Implement Model Context Protocol integrations where standardized tool and data access provides value.
AWS Cloud and Platform Engineering
Design secure, scalable, resilient, and cost-effective AI workloads on AWS.
Build containerized or serverless solutions using Amazon ECS, AWS Fargate, Amazon EKS, AWS Lambda, Amazon API Gateway, AWS Step Functions, and Amazon EventBridge.
Implement secure networking, identity, encryption, secrets management, and least-privilege access using Amazon VPC, AWS IAM, AWS KMS, and AWS Secrets Manager.
Provision infrastructure using Terraform, AWS CDK, or AWS CloudFormation.
Design for high availability, disaster recovery, workload isolation, scalability, and graceful degradation
Any Graduate
No related jobs found
← Back to jobs