Machine Learning Engineer - Captions

Salary: $170K - $230K Employment Type: Full-Time Location: New York City, NY (Union Square HQ - In-person required)

---

Position Overview

Captions is the leading video AI company, building the future of video creation. We are seeking a talented Machine Learning Engineer to partner closely with our Researchers and bring large-scale multimodal video diffusion models into production. This role involves optimizing and deploying cutting-edge generative models (tens to hundreds of billions of parameters) to achieve low-latency, high-throughput inference at scale. You will work on advanced AI technologies including audio-video generation, diffusion architectures, and temporal modeling, impacting millions of creators globally.

---

About Captions

Captions is a rapidly growing team of ambitious, experienced, and devoted engineers, researchers, designers, marketers, and operators based in NYC. We are on a mission to serve the next billion video creators. As an early team member, you will have an outsized impact on our product and company culture.

We are backed by top-tier investors and entrepreneurs, including: Index Ventures, Kleiner Perkins, Sequoia Capital, Andreessen Horowitz, Uncommon Projects, Kevin Systrom, Mike Krieger, Lenny Rachitsky, Antoine Martin, Julie Zhuo, Ben Rubin, Jaren Glover, SVAngel, 20VC, Ludlow Ventures, Chapter One, and more.

Recent Recognition:

• The Information: 50 Most Promising Startups
• Fast Company: Next Big Things in Tech
• The New York Times: When A.I. Bridged a Language Gap, They Fell in Love
• Business Insider: 34 most promising AI startups
• Time: The Best Inventions of 2024

---

Responsibilities

Inference & Deployment

• Develop high-performance GPU-based inference pipelines for large multimodal diffusion models.
• Build, optimize, and maintain serving infrastructure for low-latency predictions at large scale.
• Collaborate with DevOps teams on containerizing models, managing autoscaling, and ensuring uptime SLAs.

Model Optimization & Fine-Tuning

• Utilize techniques like quantization, pruning, and distillation to reduce latency and memory footprint while maintaining quality.
• Implement continuous fine-tuning workflows to adapt models based on real-world data and feedback.

Production MLOps

• Design and maintain automated CI/CD pipelines for model deployment, versioning, and rollback.
• Implement robust monitoring (latency, throughput, concept drift) and alerting for critical production systems.

Performance & Scaling

• Explore and implement cutting-edge GPU acceleration frameworks (e.g., TensorRT, Triton, TorchServe) to improve throughput and reduce costs.

---

Requirements

Technical Expertise

• Proven experience deploying deep learning models on GPU-based infrastructure (NVIDIA GPUs, CUDA, TensorRT, etc.).
• Strong knowledge of containerization (Docker, Kubernetes) and microservice architectures for ML model serving.
• Proficiency in Python and at least one deep learning framework (PyTorch, TensorFlow).

Model Optimization

• Familiarity with compression techniques (quantization, pruning, distillation) for large-scale models.
• Experience profiling and optimizing model inference (batching, concurrency, hardware utilization).

Infrastructure

• Hands-on experience with ML pipeline orchestration (Airflow, Kubeflow, Argo) and automated CI/CD for ML.
• Strong grasp of logging, monitoring, and alerting tools (Prometheus, Grafana, etc.) in distributed systems.

Domain Experience

• Exposure to diffusion models, multimodal video generation, or large-scale generative architectures.
• Experience with distributed training frameworks (FSDP, DeepSpeed, Megatron-LM).

---

Application Instructions

Please note that all roles require in-person presence at our NYC HQ. We do not work with third-party recruiting agencies; please do not contact us.