Generative AI Engineer

Key Responsibilities

Prompt engineering and optimization:

Design, test, and refine system prompts, prompt templates, and few-shot examples to improve response accuracy, relevance, consistency, and instruction following.

Agentic AI development:

Build AI agents that support planning, tool use, memory, and multi-step task execution, with appropriate controls for failures and human intervention.

MCP integration:

Develop and integrate Model Context Protocol (MCP) servers and tools to enable secure interaction with enterprise APIs, applications, and data sources.

Agent architecture:

Design tool-calling workflows, retrieval mechanisms, session memory, and context management strategies.

RAG and knowledge grounding:

Implement retrieval-augmented generation workflows, including document processing, embeddings, semantic search, and grounded responses.

Evaluation and testing:

Create evaluation datasets and automated tests to measure response quality, retrieval accuracy, task completion, tool selection, and agent reliability.

Security and responsible AI:

Implement safeguards against prompt injection, sensitive-data exposure, hallucinations, and unauthorized tool execution.

Production operations:

Deploy, monitor, troubleshoot, and improve AI applications, balancing quality, latency, reliability, and cost.

Cross-functional collaboration:

Work with product managers, architects, and engineering teams to translate business requirements into practical AI solutions.

Technical documentation:

Contribute to architecture reviews, design documentation, and engineering standards.

Required Skills and Experience

Hands-on experience building AI applications using LLMs such as GPT, Claude, Gemini, Llama, or comparable models.

Strong prompt engineering skills, including system instructions, structured outputs, few-shot prompting, and iterative response optimization.

Proficiency in Python, JavaScript/TypeScript, or Java, with experience integrating APIs and backend services.

Practical experience with RAG, embeddings, vector databases, semantic search, and enterprise knowledge retrieval.

Experience designing agent workflows, tool-calling integrations, and context management.

Ability to evaluate and improve AI responses for accuracy, relevance, completeness, tone, and adherence to instructions.

Understanding of hallucination mitigation, prompt injection risks, guardrails, and responsible AI practices.

Knowledge of secure healthcare-data handling, including HIPAA-aligned PHI/PII practices, data minimization, de-identification, masking, access controls, and audit logging.

Strong debugging, problem-solving, and communication skills.

Preferred Qualifications

Experience developing MCP servers and integrating MCP-enabled tools.

Experience integrating AI applications with healthcare platforms, electronic health records, or other regulated enterprise systems.

Experience deploying and operating AI applications on cloud platforms.

Familiarity with LLM evaluation, observability, and automated regression testing tools.

Experience implementing human review, approval workflows, and recovery mechanisms for AI agents.