AI Agent Developer: LLM Workflows & Production Systems

Willis Towers Watson; Ciudad de México; Presencial; MXN 900.000 - 1.500.000; Jornada completa

Descripción de la vacante

Willis Towers Watson seeks an AI-enabled backend engineer to design reliable, scalable solutions. You will craft multi-step agent workflows, integrate Claude/OpenAI-like LLMs with enterprise APIs, and ensure production readiness with retries, observability, and robust error handling.

You will develop agent memory via vector stores, manage tool calls, and implement human-in-the-loop controls. Collaboration with product teams and domain experts is essential to translate requirements into solid

Formación

• 2+ years of experience building backend or service-based systems using C# and/or Python.
• 1 year+ hands-on experience with large language models, prompt engineering, or AI-enabled systems.
• Strong analytical and debugging skills spanning APIs, data flows, and AI behavior.
• Ability to communicate technical decisions and trade-offs to technical and non-technical stakeholders.
• Proven ability to learn in a rapidly evolving tech environment.

Responsabilidades

• Design, build and test multi-step agent workflows using established patterns.
• Integrate LLMs with enterprise APIs and internal systems with robust retries and edge-case handling.
• Implement tool calling, function orchestration, and compensating actions for stability.
• Develop human-in-the-loop controls including approvals and escalations.
• Collaborate with product managers and domain experts to translate requirements into designs.
• Maintain and version prompt logic and supporting documentation.
• Build agent memory and retrieval using vector databases and RAG.
• Deploy AI services to production and monitor logs, traces, metrics.
• Contribute to test coverage and incident response with QA/ops teams.
• Support continuous improvement of system reliability and performance.

Conocimientos

Backend development

Python

C#

LLM integration

Prompt engineering

Azure

Observability

API integration

Debugging

Communication

Herramientas

Azure AI Foundry

LangChain

LangGraph

Pinecone

Descripción del empleo

Willis Towers Watson seeks an AI-enabled backend engineer to design reliable, scalable solutions. You will craft multi-step agent workflows, integrate Claude/OpenAI-like LLMs with enterprise APIs, and ensure production readiness with retries, observability, and robust error handling.

You will develop agent memory via vector stores, manage tool calls, and implement human-in-the-loop controls. Collaboration with product teams and domain experts is essential to translate requirements into solid