Backend Go Developer for AI Sales Product Team

We are creating a new generation of AI solutions for the sales domain. Our goal is to develop an intelligent system that will significantly increase the efficiency of commercial teams. The product is based on advanced large language models (LLMs). They do not work like ordinary chatbots, but as complex decision-making mechanisms and can autonomously manage dialogues with clients.

This is a project within the Yandex ecosystem. We gain access to the unique technologies of the ML platform and infrastructure, while remaining a flexible and ambitious team. You will join the core backend development team and solve problems at the intersection of high performance, machine learning, and business logic.

Technology stack: main language — Go; in the ML field, we use large language models (LLM), speech technologies (ASR/TTS), embeddings; in infrastructure — Kubernetes (k8s), Docker, PostgreSQL, message brokers (NATS/Kafka), Redis.

What tasks await you

Developing the AI Agent Core You will build from scratch high-load Go services for the intelligent agent: manage dialogue state, make decisions based on LLMs, and integrate with external systems.

Optimizing the RAG Platform You will work on systems for searching and supplying relevant information for AI models: implement vector and hybrid search, work with knowledge bases and metadata to make the agent's responses accurate and contextual.

Integration with the Sales Ecosystem You will design fault-tolerant APIs and interaction mechanisms with CRM systems, communication services (messengers, telephony), and other corporate tools.

Developing Pipelines In close collaboration with ML engineers, you will create and optimize pipelines for analyzing dialogues, extracting entities, and generating context-dependent actions.

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

We expect that you

• Have developed and designed web services in Go
• Deeply understand the principles of operation and scaling of relational DBMS (e.g., PostgreSQL)
• Have built high-load and fault-tolerant systems
• Can independently decompose complex tasks and are not afraid to take responsibility for key architectural components

Will be a plus

• Understand the principles of RAG systems and agent frameworks
• Are familiar with the basics of machine learning and large language models (LLM)
• Have worked with event-driven architecture and message queues
• Are interested in the sales domain and business process automation