Quanterian
Website:
quanterian.com
Job details:
About the role
We're hiring a senior full stack engineer to architect, build, and scale a production-grade AI platform built on the MERN stack, with deep retrieval-augmented generation (RAG) and agentic AI components at its core.
You'll work across the entire stack — from React front-ends, through Node.js and Express APIs, to MongoDB and vector data layers — and own the integration of large language models, retrieval systems, and multi-agent orchestration that powers the product.
This is a senior, hands-on engineering role for someone who has shipped real RAG and LLM-powered systems in production and is fluent in both modern web architecture and the AI tooling ecosystem.
Tech stack you'll work with
- Frontend: React.js (Hooks, functional components), modern state management (Redux, Zustand, or RTK Query), client-side routing, responsive UI.
- Backend: Node.js, Express.js, RESTful API design (GraphQL a plus), authentication / authorization, caching, logging.
- Database: MongoDB — schema design, indexing, aggregation pipelines, query optimization.
- AI / LLM layer: OpenAI, Anthropic, Google, and open-source models (Llama, Mistral, Cohere) integrated behind a unified provider-routing layer.
- Retrieval: Vector databases such as Pinecone, Weaviate, Qdrant, or pgvector; embedding-based search; chunking and reranking strategies.
- Agentic patterns: Tool use, multi-agent orchestration, prompt flow management, context-window handling, guardrails and safety mechanisms.
- Infrastructure: AWS, GCP, or Azure; Docker; CI/CD (GitHub Actions, GitLab CI, or similar); container orchestration a plus.
- Security: End-to-end encryption, SSO, role-based access control, audit logging, secrets and API-key management, rate limiting.
What you'll be responsible for
- Designing and building end-to-end RAG pipelines — document ingestion, chunking, embeddings, indexing, retrieval strategies, and reranking.
- Building multi-agent orchestration logic — tool use, context management, prompt flows, and guardrails for LLM-based features.
- Implementing a multi-model routing layer that selects the appropriate LLM per task and manages cost, latency, and quality trade-offs.
- Developing robust Node.js and Express APIs to coordinate retrieval and generation workflows.
- Building responsive, performant React interfaces for AI-driven user experiences.
- Designing for scalability, observability, and security across the full stack.
- Establishing engineering best practices — unit, integration, and end-to-end testing; code review; CI/CD discipline.
- Mentoring junior and mid-level engineers and contributing to architectural decisions.
Required skills
- Bachelor's or Master's in Computer Science, Engineering, or equivalent practical experience.
- 2–3 years as a full stack engineer, with at least 1 years focused on the MERN stack.
- Strong production experience building RAG-based applications — not just prototypes or proofs of concept.
- Hands-on integration of LLMs with vector / embedding-based retrieval systems.
- Working knowledge of agentic patterns — tool calling, multi-agent coordination, context-window management, and guardrails.
- Strong API design skills (REST; GraphQL a plus).
- Cloud deployment and containerization experience (Docker; Kubernetes a plus).
- Solid grasp of authentication, caching, logging, observability, and security for AI-integrated web applications.
- Strong debugging, performance-tuning, and problem-solving skills across the stack.
- Clear communication and the ability to collaborate with product, data, and AI/ML teammates.
Bonus points
- Hands-on experience with prompt engineering, LLM evaluation, and quality measurement for retrieval and generation.
- MLOps and observability for AI systems — latency, cost, drift, and retrieval-quality monitoring.
- Event-driven or microservices architectures (Kafka, RabbitMQ, or pub/sub) for AI workloads.
- ETL / ELT data pipelines for preparing and maintaining knowledge bases.
- Familiarity with MLOps tooling (MLflow, Kubeflow, SageMaker, Vertex AI, or similar).
- Open-source contributions in the RAG, agent, LLM, or MERN ecosystems.
Click on Apply to know more.