Tech Stack

Description

You will design and build components of a multi-tenant inference platform integrated with MongoDB Atlas, supporting semantic search and hybrid retrieval. You'll collaborate with AI engineers and researchers to productionize inference for embedding models and rerankers, and contribute to platform capabilities like latency-aware routing and observability.

Requirements

• 2+ years experience building backend or infrastructure systems at scale
• Strong software engineering skills in Go, Rust, Python, or C++ with performance focus
• Experienced in cloud-native architectures, distributed systems, and multi-tenant service design
• Familiar with ML model serving and inference runtimes concepts
• Knowledge of vector search systems (Faiss, HNSW, ScaNN) is a plus

Responsibilities

• Design and build multi-tenant inference platform integrated with MongoDB Atlas
• Productionize inference for embedding models and rerankers for batch and real-time use cases
• Contribute to latency-aware routing, model versioning, health monitoring, and observability
• Improve performance, autoscaling, GPU utilization, and resource efficiency in cloud-native environment
• Work across product, infrastructure, and ML teams to meet scale, reliability, and latency demands