Mid-Level AI Data & Orchestration Engineer

About the Position

We are looking for an enthusiastic Mid-Level AI Data & Orchestration Engineer to design and scale the intelligent data layers and multi-agent systems powering our platform.

This is a heavy backend and data-centric role. The core of our AI logic, orchestration, and graph-based memory systems is built using Python. You will be responsible for building complex, stateful multi-agent workflows, managing advanced Retrieval-Augmented Generation (RAG) pipelines, and engineering high-throughput ingestion pipelines. Alongside Python, you will utilize Node.js to integrate these AI services into our broader application backend and handle asynchronous workflows.

Key Responsibilities

·      Agentic Graph Systems: Architect, deploy, and optimize complex, cyclic multi-agent workflows and stateful decision loops using Python (LangGraph, CrewAI, or AutoGen).

·      Advanced RAG & Data Engineering: Design and build semantic data pipelines. Implement advanced chunking strategies, hybrid search (keyword + vector), and metadata filtering to maximize retrieval precision.

·      Vector DB Management: Build automated data ingestion pipelines in Python to clean, parse, embed, and index large scale unstructured data into Vector Databases.

·      Backend Co-existence (Node.js): Write clean, async services in Node.js to bridge our core Python AI orchestration engines with our application layers, managing webhooks and API routing.

·      AI Observability & Evaluation: Implement logging, tracking, and evaluation frameworks (e.g., LangSmith, Phoenix, or LangFuse) to debug agent trajectories, track token budgets, and minimize latency.

·      Light Client Integration: Coordinate with front-end components occasionally to ensure token-streaming performance and handle WebSocket states gracefully.

Requirements (What You Bring)

·      Experience: 3 to 5 years of professional software engineering experience, with a heavy emphasis on backend and data pipelines.

·      Python Mastery: Strong production experience writing clean, scalable, object-oriented Python (FastAPI, Pydantic) for data processing or machine learning operations.

·      AI Framework Fluency: Hands-on experience building production workflows using LangChain and LangGraph (or equivalent Python state-machine/agent frameworks).

·      Data Layer Expertise: Deep familiarity with Vector Databases (e.g., Pinecone, Qdrant, Milvus, Chroma) and traditional relational/NoSQL databases.

·      Node.js Literacy: Solid experience writing JavaScript/Node.js for microservices, API endpoints, or async event loops.

·      Core AI Concepts: Practical understanding of vector embeddings, distance metrics, reranking models, and context-window optimization.

Advantages (Nice to Have)

·      Experience with data orchestration tools like Apache Airflow, Prefect, or Dagster.

·      Familiarity with creating custom servers using the Model Context Protocol (MCP).

·      Basic experience with Vue.js/Nuxt.js or React for building quick internal administrative playgrounds/testing tools.

• ·      Model Context Protocol (MCP): Hands-on experience building, extending, or integrating custom servers and clients using MCP to connect LLMs to data sources and tools.