MLOps Lead at Yandex Crowd

Yandex Crowd is a major infrastructure service of Yandex. We use crowdsourcing to enhance business processes: data labeling, content moderation, field research, testing. We also develop internal functions for our products: customer service, telemarketing, localization, and documentation.

Our goal is to automate the routine tasks of crowd workers. We work in two directions: assessor labeling and customer service. We create and implement generative chatbots, agents, co-pilots for operators and assessors, and GenAI-based labeling. The number of our models and microservices integrated into platforms is growing. We need to maintain the stability, transparency, and uniformity of development and production deployment.

We are looking for an MLOps Lead who will assemble a team and build the infrastructure and processes for our ambitious ML solutions.

What tasks await you

Automation of ML processes You will create efficient automated pipelines for training, testing, and deploying ML models, implement CI/CD and modern ML process management tools.

Model stability and availability You will need to ensure the stable operation of models in production, monitor their status, and guarantee that services are available to users and fault-tolerant.

Development of the ML platform You will define the technical vision for the development of the ML platform at Yandex Crowd and tools for Data Scientists, establish processes to quickly and efficiently launch new solutions, reduce time-to-market, and scale best practices of the internal infrastructure.

Development of the MLOps engineering team You will form and develop the team of MLOps engineers: hire employees, mentor them, transfer expert knowledge, distribute tasks, and contribute to their professional growth.

We expect that you

• Have worked in a team on Python development
• Have led the creation and maintenance of backend services
• Have organized and implemented monitoring and logging systems
• Are proficient in working with Docker and Kubernetes
• Understand the full lifecycle of ML models and their deployment to production
• Are capable of making architectural decisions

It will be a plus

• Have led the building or evolution of CI/CD processes
• Have implemented tools for managing GenAI solutions: RAG, agents, LLM pipelines
• Have been involved in building internal ML platforms or complex ML product tools