Senior AI Engineer, Voice & Agentic Systems

FIS is a fintech company advancing the way the world pays, banks, and invests. FIS is seeking a Senior AI Engineer to architect and deliver enterprise-scale Agentic AI voice systems for banking, integrating LLMs, real-time speech interaction, secure APIs, and cloud-native architectures. The role also focuses on optimization, evaluation, observability, security, troubleshooting, cross-functional collaboration, and mentoring engineers.

Responsibilities:

• Architect and deliver enterprise-scale Agentic voice AI systems that combine deterministic workflow control, LLM-driven reasoning, tool orchestration, memory management, and secure API integration across regulated banking use cases
• Build production-grade Agentic AI solutions that integrate speech-to-text, text-to-speech, LLMs, cognitive services, and backend enterprise systems to deliver natural, reliable, real-time customer interactions
• Design agentic workflows that enable AI systems to complete complex, multi-step tasks using structured planning, function calling, tool routing, state management, and human-in-the-loop escalation where appropriate
• Optimize real-time voice interactions for low latency, reliability, scalability, and cost efficiency through streaming architecture, token optimization, semantic caching, prompt tuning, and performance engineering
• Deploy highly available AI services using cloud-native architectures across Azure and AWS, including Kubernetes, container platforms, serverless services, multi-region resilience, and secure DevOps practices
• Build evaluation, experimentation, and observability frameworks to measure conversational quality, routing accuracy, latency, containment, hallucination risk, tool-call reliability, and production performance
• Implement security, privacy, Responsible AI, and compliance guardrails to protect sensitive customer data and ensure AI systems operate within financial-services regulatory expectations
• Troubleshoot complex issues across distributed AI systems, including network latency, model behavior, prompt failures, token-limit breaches, API errors, orchestration defects, and non-deterministic agent outcomes
• Collaborate with product, architecture, engineering, cloud, data, security, and business teams to translate business requirements into scalable AI solutions
• Mentor engineers in AI-first software engineering practices, including reusable component design, automated testing, defensive programming, observability, deployment discipline, and production support