<p style="text-align:left">Cambium Learning® Group is an award-winning educational technology solutions leader dedicated to helping all students reach their potential through individualized and differentiated instruction. Using a research-based, personalized approach, <a href="https://himalayas.app/companies/cambium-learning-group">Cambium Learning Group</a> delivers SaaS resources and instructional products that engage students and support teachers in fun, positive, safe and scalable environments. These solutions are provided through Learning A-Z® (online differentiated instruction for elementary school reading, writing and science), ExploreLearning® (online interactive math and science simulations, a math fact fluency solution, and a K–2 science solution), Voyager Sopris Learning® (blended solutions that accelerate struggling learners to achieve in literacy and math and professional development for teachers), and VKidz Learning (online comprehensive homeschool education and programs for literacy and science). We believe that every student has unlimited potential, that teachers matter, and that data, instruction, and practice are the keys to success in the classroom and beyond.</p><h3>Job Overview:</h3><p>We are seeking a talented Machine Learning Engineer II to join our CAI machine learning and scoring development team. In this role, you will be the crucial bridge between applied research and production systems. Working alongside a cross‑functional group of mathematicians, computer scientists, psychometricians, and statisticians, you will design and deploy custom machine learning solutions for our clients and internal platforms.</p><p>The ideal candidate is a full‑stack ML practitioner who is equally comfortable discussing algorithmic design with researchers and architecting scalable, low‑latency production systems. You will own the full software development lifecycle—transforming research prototypes into optimized, production‑ready solutions using modern AWS infrastructure such as SageMaker, ECS, and Lambda, with an emphasis on high‑throughput inference and PyTorch‑to‑ONNX model optimization.</p><h3>Job Responsibilities:</h3><ul><li><b>Full-Lifecycle ML Development:</b> Lead the transition of machine learning models from theoretical prototypes into scalable, high-performance production systems.</li><li><b>AWS Cloud Architecture &amp; Deployment:</b> Architect and deploy ML solutions utilizing <b>AWS ECS</b> (Elastic Container Service) for containerized workloads and <b>AWS Lambda</b> for serverless, event-driven inference pipelines.</li><li><b>Model &amp; Inference Optimization:</b> Optimize PyTorch models for production deployment by converting them to ONNX formats. Apply advanced inference optimization techniques (quantization, pruning, ONNX Runtime) and memory-efficient attention mechanisms like <b>Flash Attention</b> to minimize latency and maximize throughput.</li><li><b>Infrastructure &amp; Engineering Best Practices:</b> Champion infras