Loading...
SeniorFull-timePythonRustC++AWSAzureKubernetesMongoDBGoAIML
Salary Range
$132,600 - $179,400 /year
This salary is estimated based on similar roles. The actual salary may vary.
About the Role
We’re looking for a Senior Engineer to help build the next-generation inference platform that supports embedding models used for semantic search, retrieval, and AI-native experiences in MongoDB Atlas. You’ll join the broader Search and AI Platform organization and collaborate with ML researchers and engineers from our Voyage.ai acquisition.
Key Responsibilities
- →Design and build components of a multi-tenant inference platform integrated directly with MongoDB Atlas, supporting semantic search and hybrid retrieval
- →Collaborate with AI engineers and researchers to productionize inference for embedding models and rerankers — enabling both batch and real-time use cases
- →Contribute to platform capabilities such as latency-aware routing, model versioning, health monitoring, and observability
- →Improve performance, autoscaling, GPU utilization, and resource efficiency in a cloud-native environment
- →Work across product, infrastructure, and ML teams to ensure the inference platform meets the scale, reliability, and latency demands of Atlas users
- →Gain hands-on experience with tools like vLLM and container orchestration with Kubernetes
Requirements
- ✓5+ years of experience building backend or infrastructure systems at scale
- ✓Strong software engineering skills in languages such as Go, Rust, Python, or C++, with an emphasis on performance and reliability
- ✓Experienced in cloud-native architectures, distributed systems, and multi-tenant service design
- ✓Familiar with concepts in ML model serving and inference runtimes, even if not directly deploying models
- ✓Knowledge of vector search systems (e.g., Faiss, HNSW, ScaNN) is a plus
- ✓Comfortable working across functional teams, including ML researchers, backend engineers, and platform teams
- ✓Motivated to work on systems integrated into MongoDB Atlas and used by thousands of developers
