Lead AI Software Engineer — Scalable Backend Orchestration

Refinitiv; Ciudad de México; Híbrido; MXN 1.000.000 - 1.600.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Hybrid work model

Career development programs

Two mental health days off

Headspace app access

Tuition reimbursement

Volunteer days

Descripción de la vacante

Thomson Reuters is seeking a Lead Software Engineer AI (Staff Engineer) to own backend AI orchestration at scale, shaping MCP servers, agents, and novel LLM usage. You will drive architecture, mentor engineers, and collaborate with AI Labs and product teams to deliver reliable, auditable workflows for accountants.

You will design end-to-end backend services (C#/.NET, Python, FastAPI, PostgreSQL, AWS) and build scalable AI orchestration, including multicall strategies, guardrails, and multi‑step

Formación

• Bachelor's degree or equivalent experience in CS/engineering.
• 10+ years total tech experience with 2+ years in a lead role and 5+ years in full-stack development.
• Experience shipping production AI systems in enterprise environments.

Responsabilidades

• Design and own AI-first backend systems and orchestrations.
• Lead multi-quarter initiatives across AI, product, and infra.
• Mentor engineers and drive quality, architecture, and testing practices.

Conocimientos

AI systems

Distributed systems

Leadership

Mentoring

C#/.NET

Educación

Bachelor's degree in Computer Science/Engineering or related field

5+ years full-stack development

Herramientas

ASP.NET Core

PostgreSQL

AWS

React/JS frameworks

Descripción del empleo

Thomson Reuters is seeking a Lead Software Engineer AI (Staff Engineer) to own backend AI orchestration at scale, shaping MCP servers, agents, and novel LLM usage. You will drive architecture, mentor engineers, and collaborate with AI Labs and product teams to deliver reliable, auditable workflows for accountants.

You will design end-to-end backend services (C#/.NET, Python, FastAPI, PostgreSQL, AWS) and build scalable AI orchestration, including multicall strategies, guardrails, and multi‑step