Company:
MMC Corporate
Description:
AI Applications Developer
We are seeking an AI Applications Developer to join our team at Marsh. This role will be based in Mexico City. This is a hybrid role that has a requirement of working at least three days a week in the office. As an AI Applications Developer at Marsh, you will be design, build, and deploy production-grade AI applications that automate and enhance finance operations. You’ll work end-to-end from solution design to deployment and monitoring. Candidates from diverse backgrounds are welcome; proven success delivering ML and DS projects and LLM-powered applications matters most.
We will count on you to:
- AI App Delivery: Architect and deploy LLM-powered applications with secure backends and integrations.
- LLM & Agents: Use LLM APIs (OpenAI, Azure OpenAI, Anthropic, Google).
- Implement tool/function calling, multi-turn flows, guardrails, and error recovery using agent frameworks.
- Prompt Engineering: Design system prompts, tool schemas, guardrails; maintain prompt versioning.
- Memory & Context: Implement conversation/session memory, summarization/distillation, caching, and cost-aware context strategies.
- Evaluations: Define golden sets and automated LLM evaluations (grounding, hallucinations, latency, cost), plus human-in-the-loop gates.
- MLOps & Ops: CI/CD, monitoring (quality, latency, cost), tracing/observability, rollback/fallbacks; enforce security, compliance, and data privacy.
- Collaboration & Documentation: Translate business needs into technical designs; write clear documentation and mentor peers.
What you need to have:
- Must have real production experience deploying Agentic Applications.
- 5+ years of experience delivering machine learning and data science projects in production, including modeling, deployment, and monitoring.
- Strong Python skills for application and service development; solid API design experience.
- Hands-on experience with LLM APIs, prompt engineering, and agent frameworks (extra points for in-depth LangChain/LangGraph experience).
- Cloud experience (Azure or AWS) for secure deployments (serverless or containerized), secrets/configuration, IAM.
- Proficiency with Git/GitHub and SQL; rigorous data handling and privacy practices.
- Excellent communication skills in English (C1).
What makes you stand out:
- Orchestration (Airflow/Dagster), feature stores, event buses.
- Kubernetes/Docker, API gateways, OAuth.
- Finance domain exposure (forecasting, anomalies, reconciliation, commentary) useful context, not mandatory.
Why join our team:
- We help you be your best through professional development opportunities, interesting work and supportive leaders.
- We foster a vibrant and inclusive culture where you can work with talented colleagues to create new solutions and have impact for colleagues, clients and communities.
- Our scale enables us to provide a range of career opportunities, as well as benefits and rewards to enhance your well-being.
Marsh McLennan (NYSE: MMC) is a global leader in risk, strategy and people, advising clients in 130 countries across four businesses: Marsh, Guy Carpenter, Mercer and Oliver Wyman. With annual revenue of $24 billion and more than 90,000 colleagues, Marsh McLennan helps build the confidence to thrive through the power of perspective. For more information, visit marshmclennan.com, or follow on LinkedIn and X.
Marsh McLennan is committed to creating a diverse, inclusive and flexible work environment. We aim to attract and retain the best people and embrace diversity of age, background, disability, ethnic origin, family duties, gender orientation or expression, marital status, nationality, parental status, personal or social status, political affiliation, race, religion and beliefs, sex/gender, sexual orientation or expression, skin color, or any other characteristic protected by applicable law.
Marsh McLennan is committed to hybrid work, which includes the flexibility of working remotely and the collaboration, connections and professional development benefits of working together in the office. All Marsh McLennan colleagues are expected to be in their local office or working onsite with clients at least three days per week. Office-based teams will identify at least one “anchor day” per week on which their full team will be together in person.