Applied ML Data Mining Engineer — Retrieval & Datasets

Socket.dev; Santa Clara (CA); On-site; USD 140,000 - 220,000; Full time

Job summary

Socket.dev is seeking a Software Engineer or Senior Software Engineer to advance data-centric ML for autonomous-driving datasets. You will build mining, retrieval, and ranking methods, and ship end-to-end search experiences with Python APIs, TypeScript front-end, and scalable pipelines.

Your work will advance vector search, embeddings, and scalable indexing over fleet-scale data, contributing to QMS compliance and production reliability.

Qualifications

• BS/MS/PhD in CS/ML or related field, or equivalent practical experience.
• Experience with ML/data-mining methods and evaluation with PyTorch or TensorFlow.
• Understanding of embeddings, similarity search, and data-centric ML.
• Strong ownership and ability to drive projects end-to-end.

Responsibilities

• Develop and evaluate mining, retrieval, and ranking methods using signals like embeddings and temporal behavior.
• Build and evolve semantic image/video/scenario search with vector search and filters.
• Operate distributed mining, inference, and indexing pipelines over fleet-scale data.
• Design and ship mining products end to end: Python APIs, services, and modern frontend experiences.
• Ensure work aligns with the company’s QMS requirements and contributes to continuous improvement.

Skills

PyTorch or TensorFlow

Embeddings & similarity search

Model evaluation metrics

Ownership / self-driven

Education

BS/MS/PhD in CS/ML or related field

Tools

Python

TypeScript/React

SQL

REST APIs

Milvus/FAISS/pgvector

Job description

Socket.dev is seeking a Software Engineer or Senior Software Engineer to advance data-centric ML for autonomous-driving datasets. You will build mining, retrieval, and ranking methods, and ship end-to-end search experiences with Python APIs, TypeScript front-end, and scalable pipelines.

Your work will advance vector search, embeddings, and scalable indexing over fleet-scale data, contributing to QMS compliance and production reliability.