FLEX Software Engineer, Conversational AI and Intelligent Search

Temporary FLEX Software Engineer role focused on production conversational AI and intelligent search capabilities for Marriott experiences.

Responsibilities

• Design, build, test, deploy, and operate production-grade conversational AI applications, intelligent search experiences, and AI agents.
• Create multi-turn conversational workflows including agent routing, structured tool calling, memory, and context management.
• Deliver retrieval-augmented generation and intelligent search solutions using embeddings, vector retrieval, hybrid search, reranking, and grounding techniques.
• Integrate AI solutions with internal APIs, search platforms, databases, model providers, and external services.
• Develop scalable backend services including APIs, streaming interfaces, asynchronous workflows, and event-driven processing.
• Implement prompt management and model configuration, including structured outputs, safety guardrails, and secure authentication and authorization.
• Apply secure data-handling practices while improving accuracy, grounding, relevance, personalization, latency, scalability, and cost efficiency using measurement and iterative engineering.
• Build and maintain unit, integration, contract, end-to-end, and automated regression tests for AI applications and supporting services.
• Partner with Quality Engineering to define evaluation datasets, release-quality thresholds, and repeatable AI evaluation and experimentation practices.
• Implement logging, metrics, distributed tracing, alerting, dashboards, and production diagnostics using OpenTelemetry and application-performance monitoring platforms.
• Troubleshoot complex production issues across models, prompts, application code, infrastructure, data stores, and downstream dependencies, including root-cause analysis.
• Contribute to architecture reviews, code reviews, release planning, CI/CD implementation, progressive delivery, and production support.
• Evaluate emerging generative AI technologies and make pragmatic architecture and implementation decisions aligned with business, security, reliability, and operational needs.
• Collaborate across Product, Engineering, Quality Engineering, Security, Platform, and Operations to deliver inclusive, accessible, high-traffic customer experiences.

Requirements

• Bachelor’s degree in Computer Science, Software Engineering, Information Technology, or related field, or high school diploma/GED with 4 additional years of professional experience.
• 2+ years of professional software engineering experience.
• Programming experience in Python, Java, JavaScript, TypeScript, or a comparable language; experience building and operating production backend services and APIs.
• Experience building cloud-native applications including containers, Kubernetes, distributed systems, and automated CI/CD pipelines.
• Hands-on experience delivering generative AI, conversational AI, intelligent search, retrieval-augmented generation, or agent-based applications, including model APIs, prompt engineering, tool calling, structured outputs, embeddings, retrieval, context management, testing, observability, secure development, and sensitive-data handling.

Preferred Qualifications

• Experience building multi-agent or workflow-orchestrated AI systems using frameworks such as LangChain, LlamaIndex, Semantic Kernel, AutoGen, Temporal, or comparable tools.
• Experience with vector databases, hybrid search, reranking, retrieval optimization, AI evaluation, LLM-as-judge patterns, experimentation, latency and token optimization, caching, and cost management.
• Experience with responsible AI practices including prompt-injection defenses, content safety, and red-team testing, plus observability with OpenTelemetry and application-performance monitoring; canary deployments, feature flags, progressive delivery, and localized accessible high-traffic customer applications.

Technologies

• Python, Java, JavaScript, TypeScript
• Kubernetes, CI/CD, containers
• Retrieval-augmented generation, embeddings, vector retrieval, hybrid search, reranking
• OpenTelemetry, application-performance monitoring platforms
• LangChain, LlamaIndex, Semantic Kernel, AutoGen, Temporal
• Vector databases
• LLM-as-judge patterns, caching
• Prompt-injection defenses, content safety, red-team testing
• Feature flags, progressive delivery, canary deployments

Location and Employment Details

• Location: Bethesda, MD, United States (remote)
• Full-time schedule
• Located Remotely: Y
• Job Number: 26123873
• Job Category: Information Technology
• Location Address: 7750 Wisconsin Avenue, Bethesda MD 20814, United States
• Pay Range: $41.97-$69.23 per hour
• Position Type: Management
• Temporary position: Assignment through September 11, 2027
• Expiration Date: 11/10/2026

Benefits

• 401(k) plan
• Stock purchase plan
• Discounts at Marriott properties
• Commuter benefits
• Employee assistance plan
• Childcare discounts
• Medical, dental, vision
• Health care flexible spending account
• Dependent care flexible spending account
• Life insurance
• Disability insurance
• Accident insurance
• Adoption expense reimbursements
• Paid parental leave
• Washington Applicants Only: employees accrue paid sick leave; 0.0384 PTO balance for every hour worked; eligible to receive minimum of 9 holidays annually

Marriott Headquarters (HQ) supports hybrid work for HQ-based roles, generally requiring regular in-office presence at the assigned location; remote roles are clearly identified in the posting.