Senior AI Backend (Node.js) Developer
MEDvidi is looking for a Senior Backend Node.JS Developer to join our AI Automation team and lead the development of AI-driven integrations within our telemedicine platform.
In this role, you will design and scale backend systems that connect our core product with AI services to automate workflows for patients and healthcare providers. You will work on production-grade systems with a strong focus on reliability, security, and user experience, and see your solutions directly improve how thousands of people interact with our platform.
Responsibilities:
- 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.
Requirements:
- 8+ years of commercial backend development with Node.js in product companies (preferably SaaS), with deep understanding of complex backend systems end-to-end.
- Strong experience designing backend architectures and technical solutions, including system decomposition, scalability trade-offs, long-term maintainability, and technology choices.
- Proven experience building scalable, high-performance backend systems and robust APIs, with strong understanding of data flows, failure modes, and performance bottlenecks.
- Hands-on experience integrating LLMs into real production systems (not demos), with focus on reliability, predictability, cost control, and measurable quality improvements.
- Experience designing and implementing multi-agent AI workflows in production, including orchestration logic, ReAct patterns, supervisor-agent architectures, and clear understanding of when such approaches are justified.
- Practical experience with modern AI frameworks, primarily LangGraph, and familiarity with alternatives such as Haystack, CrewAI, AutoGen, SmolAgents, or similar.
- Solid experience designing and optimizing RAG pipelines, including embeddings, vector databases, retrieval strategies, ranking approaches, chunking strategies, and quality vs latency trade-offs.
- Strong prompt engineering skills, including structured outputs (JSON/YAML), schema validation, evaluation-based iteration (e.g., AI eval frameworks), and reliability techniques beyond prompt-only solutions.
- Understanding of LLM infrastructure in production environments, including:
- tracing & observability (LangSmith, Langfuse, or similar),
- evaluation frameworks and quality metrics,
- tooling & orchestration,
- safety mechanisms and guardrails,
- performance and cost optimization.
- Hands-on experience with PostgreSQL (schema design, migrations, performance optimization).
- Hands-on experience with AWS (CDK, Lambda, API Gateway, ElastiCache).
- Good understanding of observability and debugging (Grafana, OpenSearch, or similar).
- Strong engineering culture: testing, CI/CD, and peer reviews.
