Senior Real-Time Search ML Engineer – Large-Scale Systems

Socket.dev; Cupertino (CA); On-site; USD 180,000 - 240,000; Full time

Job summary

Socket.dev in Cupertino is building a real-time search experience at massive scale, blending Generative AI with Information Retrieval. We are seeking a highly experienced Search Systems Engineer to design, develop, and optimize distributed search systems in production.

You will own indexing, ranking, embeddings in the query path, and collaborate with ML teams to deploy models with low latency. Requirements include 10+ years in ML/DS/SE, strong IR fundamentals, and proficiency in Go/Java/C++ and

Qualifications

• Bachelor's or Master's degree in Computer Science, ML, Statistics, or related field with strong emphasis on search infrastructure.
• 10+ years in ML, data science, or software engineering roles focused on search, with hands-on production experience.
• Deep understanding of information retrieval, query understanding, augmentation, and multi-stage ranking.
• Experience with real-time systems, model retraining pipelines, and distributed search stacks.

Responsibilities

• Design, build, and optimize large-scale search systems used in production.
• Develop ranking models, embeddings, and multi-objective optimization for relevance, diversity, and freshness.
• Deploy and monitor ML models in live query paths with low latency and high throughput.

Skills

Large-scale search

Information retrieval

Query understanding

Go

Java

C++

Python

PyTorch

XGBoost

Spark

Kubernetes

OpenSearch

FAISS

LLMs in production

Vector search

Streaming pipelines

Education

Bachelor's degree

Master's degree

Tools

Docker

Kubernetes

Spark

Kafka

FAISS

ONNX Runtime

TensorRT

TorchServe

OpenSearch

MLflow

Weights & Biases

Job description

Socket.dev in Cupertino is building a real-time search experience at massive scale, blending Generative AI with Information Retrieval. We are seeking a highly experienced Search Systems Engineer to design, develop, and optimize distributed search systems in production.

You will own indexing, ranking, embeddings in the query path, and collaborate with ML teams to deploy models with low latency. Requirements include 10+ years in ML/DS/SE, strong IR fundamentals, and proficiency in Go/Java/C++ and