Backend Developer for Alice Pro

Alice Pro is a smart assistant for personal and work productivity that can, among other things, create knowledge bases from uploaded files and links, as well as connect Yandex 360 services as sources for answers. The product is focused on B2B: we supply it to organizations under a license in two formats — SaaS (currently in Yandex infrastructure) and on-premises. In the future, we plan integration with Yandex Cloud and development of cloud capabilities.

Why you will be interested

• Influence on the architecture of a B2B product: from APIs and queues to administration, access rights, and observability.
• Work at the intersection of backend and ML: document processing pipelines, integrations, quality and stability in production.
• Participation in major product phases — in the development of the cloud version and integration with Yandex Cloud.

Technologies

• Language: Go.
• Infrastructure: YDB, YMQ, S3 Object Storage, Docker.
• ML models: LLM (Alice AI, YaGPT 5.1 Pro and others), OCR, ASR, VLM, text-embeddings.

What tasks await you

Platform for Process Automation The platform allows creating customizable automated workflows with AI. Users build data processing chains, where each node performs a specific task: analyzes documents, generates answers, extracts information. You will develop this functionality on the backend — increasing the number of integrations and platform capabilities.

Development of the B2B Platform Alice Pro is sold under licenses to other companies in a SaaS installation format. There are many ideas on how to improve the product's usability for business customers: development of the admin panel, access control, logs, statistics, etc.

Development of Infrastructure for ML Components The product is built on a set of technologies that the ML team constantly measures and improves. It is necessary to add new components, experiment, and improve the system to increase quality.

More about backend at Yandex — in the channel Yandex for Backend

We expect that you

• Develop web services in Go
• Have led projects: know how to decompose medium and large tasks and build a technical implementation plan
• Know the structure of relational DBMS and the main principles of their scaling
• Understand approaches to designing high-load services
• Can effectively prioritize in conditions of several parallel projects

Will be a plus

• Have an understanding of the architecture of RAG pipelines
• Understand the general principles of LLM operation and their limitations in production
• Show initiative in improving code and processes