AI Engineer

Entertainment Partners is building production-grade AI and agentic systems that power an intelligent product suite, with a focus on applied ML engineering, LLM product development, and production system design. In this hybrid role based in Tempe, AZ, you will help design, train, evaluate, and deploy AI/ML capabilities that integrate with EP’s enterprise data and product APIs.

What You’ll Do

• Design, develop, train, fine-tune, and evaluate machine learning models using PyTorch and related libraries such as torchvision, torchaudio, torch.nn, and torch.optim.
• Build and maintain ML training pipelines, experiment tracking workflows, and model evaluation frameworks.
• Implement transformer-based models and production LLM integrations for use cases including NLP, information extraction, classification, and generation.
• Use parameter-efficient fine-tuning approaches such as LoRA, QLoRA, and PEFT to adapt foundation models to EP domains including payroll, residuals, and production management.
• Design and implement RAG architectures using vector databases and semantic search pipelines with tools including pgvector, Pinecone, and Weaviate.
• Optimize inference for latency and throughput using quantization, batching, and caching strategies for production serving.
• Develop and maintain AI evaluation frameworks, including automated evals as unit tests, to support reliable, safe, production-grade model behavior.
• Build LLM-powered agentic workflows using LangChain and LangGraph, along with EP’s internal MCP (Model Context Protocol) server architecture.
• Create multi-step reasoning pipelines, tool-calling agents, and autonomous task execution systems that integrate with EP enterprise data and product APIs.
• Apply prompt engineering strategies such as few-shot templates, chain-of-thought scaffolding, and structured output validation.
• Apply and maintain EP AI quality engineering standards including failure taxonomy, runtime guardrails, and evidence-driven release gates.
• Contribute to EP’s Enterprise Context Engine, described as a governed, zero-data-retention AI context layer exposed via MCP to Tabnine Agent and Claude Code.
• Build and maintain MLOps infrastructure for training, experiment tracking with MLflow and Weights & Biases, model versioning, and deployment.
• Containerize and deploy ML services using Docker and Kubernetes, integrating with CI/CD pipelines such as GitHub Actions and Azure DevOps.
• Monitor model performance in production, including drift detection, feedback loops, and automated retraining triggers.
• Support security, privacy, and compliance requirements, including data minimization and access control for sensitive payroll data.
• Collaborate with data engineering on feature stores, data pipelines, and training data infrastructure.
• Work with the Chief Architect AI & Data and CAIO to define AI architecture patterns and best practices for the EP engineering organization.
• Partner with product managers, UX designers, and full stack engineers to translate AI capabilities into product features.
• Conduct code reviews for AI/ML code, emphasizing reproducibility, correctness, and production readiness.
• Mentor engineers on AI engineering fundamentals, LLM integration patterns, and responsible AI practices.
• Stay current with the AI/ML landscape and evaluate new models, frameworks, and techniques for potential application at EP.
• Contribute to EP’s PE AI Maturity Scorecard (S1–S3) by advancing organizational AI capability maturity.
• Represent EP AI engineering practices during Architecture Review Board discussions.

Required Qualifications

• Bachelor’s or Master’s degree in Computer Science, Machine Learning, Statistics, Mathematics, or a related quantitative field.
• 6–10+ years of professional software engineering experience, including 3+ years focused on ML/AI engineering in production environments.
• Expert-level proficiency in Python and deep familiarity with the Python ML/AI ecosystem.
• Hands-on production experience with PyTorch, including model definition (nn.Module), custom training loops, autograd, GPU acceleration (CUDA), and model serialization (TorchScript, ONNX).
• Experience with Hugging Face Transformers, Datasets, and PEFT, including fine-tuning and adapting foundation models.
• Demonstrated experience building RAG pipelines, including chunking strategies, embedding models, vector store selection, and retrieval evaluation.
• Production experience integrating LLM APIs (OpenAI, Anthropic, and open-source via vLLM/Ollama) and building reliable prompt engineering systems.
• Experience with LangChain or LangGraph for multi-step agents and tool-calling workflows.
• Strong grounding in ML fundamentals, including learning paradigms, loss functions, regularization, evaluation metrics, and statistical validation.
• Experience with experiment tracking tools such as MLflow, Weights & Biases, or Comet, with a focus on reproducible ML workflows.
• Working knowledge of containerization (Docker) and cloud ML services including AWS SageMaker, Azure ML, or OCI Data Science.
• Experience with SQL and NoSQL databases, including designing data pipelines for ML training and inference.

Technologies

• Python, PyTorch, torchvision, torchaudio, torch.nn, torch.optim
• LoRA, QLoRA, PEFT, RAG, pgvector, Pinecone, Weaviate
• Quantization, MLflow, Weights & Biases, Docker, Kubernetes
• GitHub Actions, Azure DevOps, LangChain, LangGraph
• MCP (Model Context Protocol), Tabnine Agent, Claude Code
• TorchScript, ONNX, Hugging Face Transformers, Hugging Face Datasets, CUDA
• OpenAI, Anthropic, vLLM, Ollama, Comet
• AWS SageMaker, Azure ML, OCI Data Science, SQL, NoSQL, vector databases
• Triton Inference Server, TorchServe, Ray Serve, TensorFlow, JAX, OpenCV, Kubeflow, KFServing
• Enterprise Context Engine, MLOps, CI/CD

Compensation and Benefits

• Salary range: USD 140,000 - 180,000 per year
• Health, Dental, and Vision options
• 401(k) retirement savings plan with company match
• Paid holidays, vacation time, and sick time
• Participation in company equity plans
• Employee Assistance Program, mental health and wellness programs
• Training and development
• Annual bonus and merit reviews

Preferred Qualifications

• Experience with additional deep learning frameworks (TensorFlow, JAX) or framework interoperability (ONNX).
• Familiarity with computer vision (torchvision, OpenCV) or speech/audio processing (torchaudio).
• Experience with model compression techniques: quantization (INT8, FP16, BF16), pruning, distillation.
• Experience serving ML models at scale using Triton Inference Server, TorchServe, Ray Serve, or similar.
• Contributions to open-source ML projects or published research (papers, patents, or technical blog posts).
• Experience with responsible AI frameworks, bias evaluation, and AI governance practices.
• Familiarity with MCP server development for exposing tools to AI agents.
• Prior domain experience in payroll, fintech, media, or enterprise SaaS environments.
• Experience with Kubernetes-based ML workload orchestration (Kubeflow, KFServing, or similar).
• Hybrid work environment familiarity: Burbank, CA headquarters with flexible remote schedule.
• On-call availability as needed for production AI system incidents and model deployment events.
• Access to GPU-accelerated compute environments (cloud-based) for training workloads.
• Ability to sit for extended periods at a computer workstation, and to use a keyboard and mouse.
• Occasional participation in early-morning or evening sessions to coordinate with distributed teams or international partners.