JOB SUMMARY
Seeking a Senior AI Engineer to design, build, and scale a production-grade Generative AI and Data Platform on AWS. This role will enable LLM-powered capabilities through vector search, graph-based knowledge systems, and governed data pipelines, owning end-to-end delivery across the AI lifecycle, including data ingestion, knowledge curation, embeddings, retrieval systems, backend services, APIs, and CI/CD pipelines. The engineer will partner with product and engineering teams to operationalize AI capabilities in externally facing applications and drive evolution toward agentic AI systems.
Key Responsibilities
1. GenAI Enablement & Integration
- Build and operationalize LLM-powered applications using RAG, embeddings pipelines, and prompt orchestration/evaluation frameworks.
- Design and implement vector search systems using Amazon OpenSearch.
- Develop graph-based knowledge systems using Amazon Neptune for relationships, lineage, and explainability.
- Integrate supporting infrastructure: Amazon ElastiCache (Redis) and DynamoDB.
- Implement agentic workflows using frameworks like LangGraph, AutoGen, or CrewAI.
- Integrate with LLM frameworks such as LangChain or LlamaIndex, focusing on tool calling, retrieval orchestration, and context management.
- Define standards for tool integration and context-sharing patterns (MCP-style designs).
- Evaluate LLM models and retrieval strategies across latency, cost, accuracy, and context limitations.
2. Data Pipelines & Knowledge Engineering
- Design and build scalable data pipelines using Databricks and Apache Spark.
- Implement data ingestion, transformation, and document processing (chunking, metadata tagging) pipelines.
- Implement embedding generation and indexing.
- Ensure high data quality standards through validation, completeness, consistency, and monitoring.
- Implement data governance frameworks including data classification, access controls, retention policies, auditability, and lineage tracking.
3. Backend Services & APIs
- Develop backend services exposing AI capabilities through secure and scalable APIs.
- Define best practices for API contracts, versioning, reliability (retry logic, circuit breakers, idempotency).
- Enable reusability of platform capabilities across teams and applications.
4. Deployment, MLOps & Operational Excellence
- Build and manage CI/CD pipelines for AI and data workloads.
- Deploy production systems using Docker and Kubernetes.
- Implement deployment strategies such as blue/green deployments, canary releases, rollback strategies, and feature flags.
- Ensure system reliability through monitoring (latency, failures, cost, data freshness), alerting, observability, secrets management, and least-privilege access.
- Optimize platform performance and cost.
5. LLM Observability, Evaluation & Quality
- Define and track GenAI quality metrics, including grounding/faithfulness, retrieval relevance, response consistency, and latency/cost per request.
- Implement prompt/version tracking, offline evaluation pipelines, and continuous improvement workflows.
6. LLM Security, Safety & Compliance
- Implement secure AI systems with access control, authentication, and data protection policies.
- Implement responsible AI guardrails.
- Ensure compliance with best practices in AI safety, data privacy, monitoring, and auditability.
Required Qualifications
- Strong experience in Generative AI / LLM systems (RAG, embeddings, prompt engineering).
- Hands-on experience with the AWS ecosystem.
- Expertise in OpenSearch (vector search), Neptune (graph databases), DynamoDB, and Redis (ElastiCache).
- Experience with LangChain / LlamaIndex.
- Experience with agentic AI frameworks (LangGraph, AutoGen, CrewAI).
- Strong programming skills, with Python preferred.
- Experience with Databricks and Apache Spark.
- Solid understanding of data pipelines, distributed systems, and API design.
- Bachelor's or Master's degree in Computer Science, Data Science, AI, or a related field.
- Proven experience building production-grade AI platforms and systems.
- Strong background in end-to-end AI/ML lifecycle delivery.
Preferred Qualifications
- Experience with model evaluation frameworks and LLM observability tools.
- Experience with AI governance and compliance frameworks.
- Experience with Kubernetes and advanced MLOps practices.
- Familiarity with Model Context Protocol (MCP) patterns and agent-based architectures.
Certifications
- None specified.
