Python Architect - Agentic AI

Key Responsibilities:

• Architect and implement enterprise-grade LLM/agent systems in Python using frameworks such as

  Autogen

  ,

  SemanticKernel

  , and

  LangGraph

  .
• Design durable agentic workflows and orchestration using

  Temporal

  (workflows, activities, signals, retries, long-running transactions).
• Build complex business software integrations (transactional flows, auditability, RBAC, idempotency) that agents operate within and act upon.
• Deploy and manage LLMs on Azure AI / Azure OpenAI: inference, scaling, cost/latency SLAs, caching, model routing.
• Design Retrieval-Augmented Generation (

  RAG

  ) pipelines, embedding stores and vector DB integrations (Elasticsearch).
• Implement safe tool/function-calling patterns and secure adapters to external systems (APIs, DBs, messaging).
• Ensure model & agent safety, hallucination mitigation, bias controls, data governance and compliance.
• Mentor teams; conduct architecture reviews, design docs and runbooks for production readiness.

Required Skills:

• Expert Python (production systems) and strong software architecture (DDD, microservices, event-driven, transactional patterns).
• Hands-on LLM/agent experience: Autogen, LangGraph, or equivalent agent frameworks.
• Proven experience building durable workflows with

  Temporal

  .
• Experience deploying/operating LLMs on

  Azure AI / Azure OpenAI

  .
• RAG, embeddings and vector DB experience.
• Familiarity with function-calling / tool integration patterns and secure API adapters.
• Strong leadership, communication and system-design skills.

Preferred Qualifications:

• Advanced degree in CS/AI/ML or related.
• Experience in NLP, RL, or production model optimization (quantization, distillation, batching).
• Experience with enterprise security, privacy and compliance for AI.