We're hiring a Senior Applied AI Engineer to build production AI systems that real customers depend on. This role is for an experienced software engineer who also understands modern AI systems. You should be comfortable building with LLMs, agents, retrieval pipelines, and workflow orchestration, but just as comfortable thinking about system design, reliability, testing, deployment, debugging, and long-term maintainability.
You'll work on everything from agent workflows and retrieval systems to backend APIs, evaluation tooling, observability, and production infrastructure. We care a lot about engineering quality. That means building systems that are understandable, testable, observable, and reliable in production. We are looking for someone who can help raise the engineering bar around AI development and bring strong technical judgment to a fast-moving environment.
What You'll Work On
Agentic AI workflows that automate complex business processes
AI-powered product experiences that combine LLMs, retrieval, backend systems, and human review workflows
Retrieval systems that connect AI agents to organization-specific knowledge and data
Backend services and APIs that allow AI systems to safely interact with internal product workflows and data
Prompting, evaluation, and observability systems that improve the quality and consistency of generated outputs
Monitoring and debugging infrastructure for production AI systems
Human-in-the-loop review systems that combine automation with expert oversight
Internal AI tooling, orchestration frameworks, and operational infrastructure
Responsibilities
Design, build, and maintain production-grade AI systems and customer-facing AI features
Develop agentic workflows using LLMs, retrieval systems, tools, APIs, and backend services
Design and implement retrieval-augmented generation (RAG) systems, including ingestion pipelines, embeddings, semantic retrieval, and context assembly
Integrate foundation models through platforms such as Amazon Bedrock or Agent Core
Develop robust prompting strategies, structured outputs, guardrails, and workflow logic for production use cases
Implement evaluation systems for prompts, agents, and workflows, including regression testing, trace review, golden datasets, and human QA processes
Monitor and improve production AI systems for quality, reliability, latency, observability, and cost efficiency
Debug AI behavior through logs, traces, evaluations, user feedback, and production telemetry
Collaborate closely with engineering, product, operations, and customer-facing teams to turn ambiguous requirements into reliable systems
Help establish strong engineering standards around testing, deployment, CI/CD, version control workflows, code review, and operational reliability
Mentor and collaborate with engineers across both software and AI disciplines
Evaluate emerging AI technologies pragmatically based on business impact, maintainability, and operational reliability
Qualifications
US Citizen or authorized to work in US
5+ years of professional software engineering experience building production systems
Strong proficiency in Python
Strong backend engineering fundamentals and experience building scalable APIs, services, distributed systems, or workflow orchestration platforms
Proven hands-on experience building and shipping AI-powered applications using LLMs, generative AI APIs, agents, retrieval systems, or related technologies in production environments
Experience designing and implementing agentic workflows, tool-calling systems, structured outputs, prompt pipelines, or retrieval-augmented generation architectures
Strong understanding of the practical challenges involved in production AI systems, including hallucination mitigation, evaluation, reliability, observability, latency, and cost management
Experience building production software systems with strong engineering standards around testing, QA, deployment, monitoring, and maintainability
Strong understanding of modern software engineering practices, including Git workflows, code review, CI/CD, automated testing, operational debugging, and release management
Experience working with cloud infrastructure, preferably AWS
Experience working with SQL and/or NoSQL databases
Strong debugging, systems-thinking, and problem-solving skills
Ability to operate effectively in fast-moving environments with evolving requirements and imperfect information
Strong communication skills and ability to collaborate across technical and non-technical teams
Preferred Qualifications
Experience with Amazon Bedrock, AWS Lambda, Step Functions, S3, DynamoDB, RDS, SQS, EventBridge, or related AWS services
Experience with LangGraph, LangChain, DSPy, Semantic Kernel, or similar orchestration frameworks
Experience building multi-step agents that interact with tools, APIs, external systems, or business workflows
Experience implementing AI evaluation systems, prompt regression testing, trace analysis, or human-in-the-loop review workflows
Experience with vector databases and semantic retrieval systems such as OpenSearch, pgvector, Pinecone, Weaviate, FAISS, or similar technologies
Experience with observability and LLMOps tooling such as LangSmith, Arize, Helicone, Weights & Biases, OpenTelemetry, or similar platforms
Experience balancing quality, latency, reliability, and cost tradeoffs in production AI systems
Experience mentoring engineers and helping establish strong engineering culture and development practices
Experience working in startup or high-ownership product environments
Ability to think critically about edge cases, failure modes, operational risk, and long-term maintainability
What Success Looks Like
AI systems that are reliable, observable, maintainable, and trusted by both customers and internal teams
Engineering practices that improve development velocity, operational quality, and long-term maintainability
AI workflows that solve meaningful business problems rather than isolated demos or experiments
Strong collaboration between product engineering and applied AI efforts
Pragmatic adoption of AI technologies based on measurable business impact and operational reliability