Python AI Backend Engineer: LLM Pipelines & Cloud Native

Centraprise; Cary (NC); On-site; USD 140,000 - 190,000; Full time

Job summary

Centraprise in Cary, NC is hiring a backend AI engineer to build cloud-native services and productionize AI/LLM pipelines. You will design Python APIs and microservices using FastAPI with async patterns and craft agentic AI workflows via LangChain/LangGraph.

You will implement embeddings, RAG, vector search, and model versioning, while packaging, serving, and monitoring models for real-time and batch workloads. Expect Kubernetes deployments and robust data engineering practices.

Qualifications

• Proficient in Python with API/microservices development.
• Experience designing scalable AI/LLM workflows and API patterns.
• Strong understanding of LLM capabilities (embeddings, RAG, vector search).

Responsibilities

• Build and productionize cloud-native backend services and AI/LLM pipelines.
• Design Python APIs and microservices (FastAPI, async patterns) for agentic workflows.
• Implement and optimize LLM features including embeddings, RAG, and model versioning.
• Package, serve, and monitor models for real-time and batch inference.
• Develop resilient, event-driven integrations and Kubernetes-based deployments.

Skills

Python

FastAPI

Async patterns

LLM workflows

LangChain/LangGraph

Embeddings

RAG

Vector search

Prompt engineering

Model versioning

Concurrency

Caching

Reliability

Rate limiting

Data engineering

Kubernetes

Helm

Azure

AKS

Performance tuning

Frontend collaboration

Model governance/monitoring

Tools

LangChain

LangGraph

Azure AKS

Helm

Job description

Centraprise in Cary, NC is hiring a backend AI engineer to build cloud-native services and productionize AI/LLM pipelines. You will design Python APIs and microservices using FastAPI with async patterns and craft agentic AI workflows via LangChain/LangGraph.

You will implement embeddings, RAG, vector search, and model versioning, while packaging, serving, and monitoring models for real-time and batch workloads. Expect Kubernetes deployments and robust data engineering practices.