Overview

GenAI Engineer - Development

Responsibilities

• 8-10 yrs (Lead), 4-5yrs (Developer)
• 8-10 yrs (Lead), 4-5yrs (Developer)

Qualifications

• Must be comfortable partnering with Nike engineers and our governance processes.
• Agentic Development/ Observability: Hands on experience with tools such as LangSmith/ Langfuse/Data Dog. Building Multi-agent services. Exposure to monitoring and tracing multi-agent/LLM workflows concepts, tool-call tracing, latency/cost/drift monitoring, Smart LLM routing, Cost Optimization and Multi-Cloud Decisioning (AWS/GCP, etc)
• Agentic Governance: guardrails and safety controls for autonomous agents, human-in-the-loop design, audit logging, and responsible AI/compliance frameworks.
• Data Strategy: experience contributing to or advising on enterprise data strategy, data architecture roadmaps, and data-driven decision-making frameworks.
• Dev Data Platforms: data pipeline design, and data engineering fundamentals is mandatory. Hands-on experience with modern data platform tooling (e.g., Databricks, Snowflake, or similar lakehouse/warehouse platforms) is a plus.
• DevOps: Exposure to CI/CD pipeline design (e.g., Jenkins, GitHub Actions, Azure DevOps), infrastructure as code (Terraform or equivalent), and automated testing/release practices.
• Strong programming skills in Python (or similar) for building GenAI applications.
• Experience with LLM application frameworks such as LangChain, LangGraph, Semantic Kernel, or AutoGen.
• Experience with RAG architectures, vector databases (e.g., Pinecone, Weaviate, FAISS, pgvector), and embeddings.
• Familiarity with prompt engineering, fine-tuning/instruction-tuning, and evaluation of LLM outputs.
• Experience integrating with major LLM providers/platforms such as OpenAI, Anthropic, Azure OpenAI, or AWS Bedrock.
• Understanding of MLOps/LLMOps practices for model deployment and lifecycle management.
• Must be comfortable partnering with Nike engineers and our governance processes.
• Agentic Observability: monitoring and tracing multi-agent/LLM workflows concepts, tool-call tracing, latency/cost/drift monitoring. Familiarity with tools such as LangSmith/ Langfuse/Data Dog, or equivalent is a plus.
• Kubernetes (K8s): deploying, scaling, and managing containerized workloads; familiarity with Helm, namespaces, and resource management. Expertise has to be strong, with decent GenAI+Data Platform exposure.
• DevOps: Strong CI/CD pipeline design (e.g., Jenkins, GitHub Actions, Azure DevOps), infrastructure as code (Terraform or equivalent), and automated testing/release practices.
• MLOPS/LLMOPs – Good exposure with model deployment and LLM agents lifecycle management. LLM production support, Endpoint provisioning and management, Model deployment and operations
• Exposure with LLM application frameworks such as LangChain, LangGraph, Semantic Kernel, or AutoGen.
• Exposure with RAG architectures, vector databases (e.g., Pinecone, Weaviate, FAISS, pgvector), and embeddings.
• Familiarity with prompt engineering, fine-tuning/instruction-tuning, and evaluation of LLM outputs.