Machine Learning Engineer

Key Responsibilities

• Design and implement end-to-end ML pipelines including data preprocessing, feature engineering, model training, evaluation, and deployment
• Build and maintain model serving infrastructure that handles real-time and batch prediction workloads with low latency and high reliability
• Develop feature stores and feature engineering pipelines to provide consistent, reusable features across training and serving environments
• Implement model monitoring, drift detection, and automated retraining pipelines to keep models performant over time
• Optimize model performance for production constraints including latency, throughput, memory, and cost
• Collaborate with data scientists to productionize research models, translating Jupyter notebook prototypes into maintainable, tested code
• Build and maintain MLOps tooling and infrastructure including experiment tracking, model registry, and CI/CD for ML

Required Skills & Experience

• 4+ years of software engineering experience with at least 2 years focused on ML systems in production
• Strong Python programming skills with emphasis on production-quality code (testing, error handling, logging, documentation)
• Experience with ML frameworks: PyTorch, TensorFlow, or JAX for model development and training
• Hands-on experience deploying models to production using serving tools like TorchServe, TensorFlow Serving, Triton, or custom APIs
• Familiarity with MLOps tools: MLflow, Weights & Biases, Kubeflow, or similar experiment tracking and pipeline orchestration
• Experience with cloud ML services (AWS SageMaker, GCP Vertex AI, or Azure ML)
• Strong understanding of Docker, Kubernetes, and microservice architectures for model deployment
• Knowledge of data processing at scale using Spark, Dask, or Ray

Nice-to-Have

• Experience with LLM fine-tuning, prompt engineering, or RAG (Retrieval-Augmented Generation) systems
• Familiarity with model optimization techniques: quantization, pruning, distillation, ONNX conversion
• Experience with feature stores (Feast, Tecton) and real-time feature serving
• Knowledge of A/B testing infrastructure for model evaluation in production
• Experience with GPU cluster management and distributed training

Tech Stack

What We Offer

• Competitive salary and equity at [Company Name]
• Access to dedicated GPU clusters and cutting-edge ML infrastructure
• Budget for ML conferences (NeurIPS, ICML, MLSys) and continuous learning
• Comprehensive health, dental, and vision insurance
• Flexible remote work with quarterly in-person team gatherings
• Opportunity to work on ML systems that serve millions of users at scale

Interview Process

1. 1Recruiter phone screen (30 min) — background, ML experience depth, and logistics
2. 2Technical phone screen (60 min) — coding exercise in Python, ML system design discussion
3. 3ML system design interview (60 min) — design an end-to-end ML pipeline for a realistic scenario, discuss trade-offs in serving, monitoring, and retraining
4. 4Coding deep-dive (60 min) — implement a data processing or model serving component with emphasis on production quality
5. 5Hiring manager and team conversation (60 min) — collaboration style, past project deep-dive, and career goals
6. 6Reference checks and offer