Generative AI Engineer

Artisan Management LLC•Asheville, NC

16h•Remote

About The Position

The Generative AI Engineer designs, builds, and ships production-grade GenAI systems for our clients and our own platforms. You will work at the intersection of large language models, agent architectures, and cloud engineering on AWS, turning ambiguous problems into reliable, secure, and measurable AI solutions. You will partner closely with delivery teams and clients to move ideas from prototype to production.

Requirements

3+ years building software in production, with 1+ year of hands-on work building LLM or generative AI applications (professional, open-source, or substantial personal projects all count).
Strong Python skills for building AI applications, services, and automation.
Practical experience with LLM application development including prompt engineering, RAG architecture, and function/tool calling.
Hands-on experience with Amazon Bedrock; familiarity with agent runtimes such as Bedrock AgentCore or comparable agent frameworks, and with agent-to-tool standards such as the Model Context Protocol (MCP) or equivalent.
Experience with at least one vector store (OpenSearch, pgvector, Pinecone, Weaviate, or similar) and embeddings.
Comfortable working in AWS day to day, including IAM, Lambda, S3, and basic networking concepts.
Experience building and consuming REST APIs and integrating LLMs into existing systems, including authentication, asynchronous patterns, streaming, and error and timeout handling. Exposure to tool and data integration for agents, such as MCP servers and clients, is a plus.
Sound judgment about handling sensitive and client data with AI systems, including PII, what may and may not be sent to a model, and basic data governance and compliance.
Able to work directly with clients and delivery teams to scope problems, run working sessions, and set realistic expectations about what generative AI can and cannot do.
Experience with Git-based workflows, containers (Docker), and shipping code through CI/CD.
Understanding of how to evaluate, guardrail, and monitor LLM systems for accuracy, safety, and cost.

Nice To Haves

Fine-tuning techniques such as LoRA, QLoRA, or PEFT, and model customization on SageMaker.
Machine learning fundamentals, feature engineering, and the model experimentation lifecycle.
Python ML and data libraries such as PyTorch, Hugging Face Transformers, scikit-learn, pandas, and NumPy.
MLOps tooling such as SageMaker Pipelines, Model Registry, or Feature Store.
Responsible AI practices including bias evaluation and explainability (for example, SageMaker Clarify).
Data engineering exposure such as SQL, AWS Glue, Athena, or Spark for preparing data that feeds AI systems.

Responsibilities

Design and build LLM-powered applications and agentic systems using Amazon Bedrock, including Bedrock AgentCore and Bedrock Agents.
Architect and implement retrieval-augmented generation (RAG) pipelines, including chunking, embeddings, vector search, and reranking.
Develop prompt strategies, tool and function calling, and orchestration using frameworks such as LangChain or LlamaIndex.
Fine-tune and customize foundation models on Amazon SageMaker when business requirements call for it.
Define and implement evaluation, guardrails, and observability so AI systems are measurable, safe, and cost-aware in production.
Deploy and operate GenAI workloads on AWS using infrastructure-as-code, containers, and CI/CD.
Collaborate with delivery teams and clients to scope solutions, run discovery, and translate requirements into working systems.
Mentor engineers on GenAI patterns and contribute to reusable accelerators and internal platforms.