Machine Learning Engineer

Yum! Brands’ is hiring Machine Learning Engineers to support the development and optimization of real-time AI systems that power the Taco Bell Voice AI experience. This role will focus on speech, natural language, and infrastructure-oriented ML tasks that help ensure performance, reliability, and adaptability of deployed voice agents.

You’ll collaborate closely with MLEs, AI Engineers, and QA to fine-tune and evaluate models, improve latency, and deliver on voice AI performance standards at drive-thru scale.

Responsibilities:

Model Development & Optimization

• Train, fine-tune, and evaluate speech recognition (ASR), LLM, and NLU models
• Contribute to model monitoring, evaluation pipelines, and test harnesses
• Optimize for speed, robustness, and market variability

Deployment & Integration Support

• Collaborate with MLEs and DevOps to validate models in staging and production
• Participate in benchmarking across markets and environments (hardware, accents, menu dynamics)
• Help investigate production issues and assist in triage/debugging

Collaboration & Experimentation

• Work with AI Engineers to deploy prompt-based systems enhanced with learned representations
• Pair with QA to improve model accuracy and regressions
• Contribute to experiment design, data selection, and annotation loops.

Mandatory Skills

• 4-8 years of experience in applied machine learning or MLE roles.
• Proficiency in Python and ML libraries such as PyTorch, TensorFlow, HuggingFace.
• Experience in fine-tuning speech, ASR, or embedding models.
• Deep experience with ML pipelines and model lifecycle management.
• Hands-on with NLP/LLM modeling,Fine-tuning (LoRA/QLoRA), Embeddings & Prompt optimization.
• Exposure on building evaluation frameworks for models.
• Experience with ML ops elements: Model tracking (MLflow or similar),Experimentation workflows, Versioning + testing,AWS experience for model training + hosting.
• Ability to work on: Model optimization,Latency/throughput tuning,Scalable inference architecture.

Preferred/good to have

• Exposure on ASR/TTS models
• Experience with: Conversation classification, Intent detection, Sentiment models.
• Prior experience with deep learning frameworks (PyTorch, TensorFlow).
• Knowledge of real-time inference systems.
• Some understanding of Voice AI orchestration.