Lead Applied ML Research Engineer

Job Duties: Acquire and manage large-scale video datasets from multiple sources, maintaining cloud-based infrastructure for storage and organization of training data at scale. Design and implement scalable data preprocessing pipelines on serverless compute infrastructure to efficiently process large volumes of video data. Develop and maintain automated data tracking and bookkeeping systems using distributed database solutions to ensure data provenance and processing status visibility. Preprocess and deliver training datasets to research teams according to specifications, iterating on preprocessing approaches based on model training feedback and requirements. Generate synthetic training data for post-training and fine-tuning workflows, establishing annotation pipelines for data labeling and quality control. Design and implement evaluation frameworks including human evaluation processes for assessing model outputs and synthetic data quality. Deploy and optimize video diffusion models on cloud GPU providers, maintaining inference scripts for improved performance, latency, and cost efficiency. Maintain and monitor model serving infrastructure, hosting API endpoints to provide production-ready model access to backend and client development teams. Ensure reliability and performance of model services through continuous monitoring and optimization of deployment infrastructure. Collaborate cross-functionally with research, engineering, and product teams to enable integration of AI models into production systems. Job Requirements: Employer will accept a Masters degree in Computer Science or related field, and 36 months of experience in experience in the job offered or in a related occupation. Position requires: 1. AWS S3 for large-scale video dataset storage; 2. DynamoDB for data tracking, provenance, and processing status management; 3. RunPod for serverless video preprocessing pipelines; 4. FFmpeg and PyAV for video manipulation and format conversion; 5. OpenCV for frame-level operations and quality assessment; 6. Kubernetes clusters for distributed training of video diffusion models and transformer architectures; 7. PyTorch for model development and experimentation; 8. Baseten for production model serving and API endpoints; 9. CUDA optimization for memory-efficient inference and chunked processing strategies; 10. Git with submodules for version control and coordinated development across research and engineering teams; 11. AWS Lambda and EC2 for automated workflow orchestration. Job ID #: CL19 The total base annual salary range for this position is $251,000 to $285,000. To apply, you must email resume to Cantina, Inc., Attn: Deborah Cardoso, VP of Finance and Team . Must reference Job ID# CL19 in subject line.