Sr AI Engineer

At T-Mobile, we invest in YOU! Our Total Rewards Package ensures that employees get the same big love we give our customers. All team members receive a competitive base salary and compensation package – this is Total Rewards. Employees enjoy multiple wealth‑building opportunities through our annual stock grant, employee stock purchase plan, 401(k), and access to free, year‑round money coaches. That’s how we’re UNSTOPPABLE for our employees! 

Are you ready to redefine the future of customer service with the power of AI? As a Sr AI Engineer on the Intent

CX team, you will drive the optimization and orchestration of AI models that power T‑Mobile’s customer service automation. Your role will focus on enhancing model outputs by leveraging the latest in prompt engineering and optimization, fine‑tuning, and agentic AI systems. You’ll design advanced tools and workflows that empower AI to think, reason, and respond in ways that feel genuinely human, tackling the most complex challenges in customer service.

Partnering with diverse teams across the company, you’ll see your innovations move swiftly from concept to reality, directly impacting customer satisfaction and setting new standards for reliability and efficiency. 

• Builds agentic AI systems that accomplish complex tasks by invoking AI models as well as internal and third‑party tools using APIs, ensuring seamless data flow in production environments
• Optimizes performance of agentic AI systems through innovative techniques such as prompt engineering, fine‑tuning and reinforcement learning using T‑Mobile’s customer interaction data
• Develops AI tools, workflows, and middleware to enhance model capabilities, such as structured reasoning, multi‑step task execution, and improved contextual memory
• Implements retrieval‑augmented generation (RAG) techniques to ensure AI responses are contextually accurate and grounded in real‑time data
• Collaborates in a highly matrixed environment with backend engineers, business experts and conversation designers to ensure AI‑driven enhancements are effectively integrated into production environments
• Tracks success metrics that aligns with business requirements and continuously evaluate and improve model quality based on those metrics
• Develops internal tooling and automation to streamline AI deployment, evaluation, and self‑improvement mechanisms
• Monitors real‑world AI performance and proactively iterates on model behavior based on live interaction data
• Stays up‑to‑date with latest LLM advancements in prompt design, prompt optimization, few‑shot learning, Tool integration protocols like MCP and AI orchestration frameworks like Agent SDK

• 4+ yrs developing and deploying machine learning models, particularly in the context of AI‑driven customer service automation
• 4+ yrs experience with advanced AI techniques such as prompt engineering, fine‑tuning, and creating AI tools and workflows
• 4+ collaborating with cross‑functional teams to integrate AI systems into production environments
• Proficiency in Python and AI development frameworks for building scalable AI applications
• Experience with operational excellence practices and observability tools (e.g., Weights & Biases, Splunk, Datadog) for monitoring, logging, and troubleshooting AI systems in production
• Experience with project management tools and agile methodologies (e.g., Jira, Azure Dev Ops) to plan, track, and deliver AI initiatives efficiently in cross‑functional environments
• Experience in LLM fine‑tuning and prompt engineering (e.g., OpenAI APIs, Hugging Face, Anthropic Claude, Google Gemini)
• Experience with AI orchestration tools (e.g., Lang Chain, Llama Index, vector databases for retrieval augmented generation)
• Hands‑on knowledge of function calling and API‑based reasoning models (e.g., using structured outputs to drive automated workflows)
• Familiarity with RAG pipelines and vector database retrieval for augmenting AI responses
• Understanding of multi‑agent architectures and best practices in agentic AI design
• Experience with real‑world AI evaluation techniques, including golden sets, synthetic data generation, and interactive testing
• Ability…