Senior AI Backend (NodeJS) Developer

MEDvidi is a leading telehealth provider specializing in mental health issues. We are building a comprehensive ecosystem of solutions and apps for mental health issues to help people achieve happiness and overall well-being.

We’re looking for a Senior Backend Developer to join our AI Automation team. This role is a unique opportunity to work at the intersection of backend engineering and applied AI, designing scalable, reliable systems that power LLM-driven features in a high-impact, product-focused environment.

What You’ll Do:

• Design and build backend architecture for AI-powered automation systems, focusing on performance, scalability, and resilience.
• Integrate and scale LLM-based product features, working closely with AI engineers and product teams.
• Implement and evolve agent-based systems (e.g., ReAct, supervisor-agent models), supporting complex decision-making flows.
• Optimize Retrieval-Augmented Generation (RAG) pipelines and LLM context enrichment using structured and unstructured data.
• Develop robust APIs for internal services and external integrations.
• Ensure strong observability, monitoring, and alerting practices using tools like Grafana and OpenSearch.
• Participate in code reviews, contribute to architecture discussions, and promote engineering excellence.
• Maintain high-quality technical documentation for newly-designed systems and APIs to ensure clarity and knowledge sharing within the team.
• Collaborate with cross-functional teams to align backend systems with product goals and user needs.

What We’re Looking For:

• 5+ years of backend development experience using Node.js.
• Solid expertise in designing and developing scalable, high-performance backend systems.
• Proficiency in Russian and English.
• Familiarity with Amazon Web Services (AWS).
• Strong knowledge of LLM integration in production, with hands-on experience building real product features, not just demos or pet projects.
• Experience implementing and tuning RAG pipelines, including knowledge of embeddings, vector databases, top_k retrieval, ranking models, etc.
• Familiarity with LLM agents and orchestration tools (ideally LangGraph, but also AutoGen, SmolAgents, etc.).
• Understanding of the ReAct paradigm, agent roles, and agent-supervisor patterns in applied use cases.
• Skilled in prompt engineering with measurable quality improvements (e.g., using AI Eval frameworks).
• Ability to work with structured outputs from LLMs (JSON/YAML), including output validation techniques beyond prompt instructions.
• Good understanding of observability tools like Grafana and OpenSearch, and experience building resilient, monitored systems.
• Proficiency in designing and developing robust APIs.
• Experience with testing practices, including unit testing, integration testing, and CI/CD automation (e.g., Jest).
• Practical experience working with PostgreSQL, including schema design and basic performance optimization.
• Product-oriented mindset, committed to delivering high-quality, user-focused solutions.
• Experience designing and building internal tools or configuration-driven systems that enable non-technical team members (e.g., analysts, product managers) to manage application logic or content.
• Skilled in prompt engineering with measurable quality improvements (e.g., using AI Eval frameworks) as a plus.