Python AI Engineer (Prompt & Agentic Systems)

About the Role

Develop end-to-end AI-enabled applications using Python, focusing on Large Language Models (LLMs) and agentic (autonomous/multi-agent) system architectures.

Responsibilities

• Design and build Python-based backend services that leverage LLMs for reasoning, extraction, summarization, and decision support.
• Engineer, version, and evaluate AI prompts/system instructions; create guardrails and optimize for reliability, latency, and cost.
• Architect and implement agentic systems, including planning, tool usage, memory, error recovery, and human-in-the-loop controls.
• Build retrieval-augmented generation (RAG) pipelines: ingest documents, chunk data, generate embeddings, and implement semantic/vector search.
• Define and monitor metrics for quality, safety, and performance of AI systems; establish observability for LLM calls.
• Integrate APIs, databases, and workflow tools to enable agent capabilities, handling authentication, rate limits, and fallbacks.
• Package and deploy services using Docker/Kubernetes; manage secrets, CI/CD pipelines, and support cost governance.
• Ensure security and compliance, including PII handling, data privacy, prompt injection defense, and audit logging.
• Collaborate cross-functionally with product, data, and security teams to deliver reliable and innovative AI features.