Machine Learning Engineer, Platform

Overview

In this role you drive end-to-end ML components for Scale GP’s enterprise Generative AI platform. You’ll shape knowledge representation, RAG pipelines, and context-aware retrieval across knowledge bases and vector stores to empower agents that deliver real impact for customers. You will own design-to-production work, collaborating with product, ML, and infrastructure teams to balance performance and cost. The role offers the opportunity to build scalable backend services and evaluation frameworks that measure retrieval quality and agent performance. You’ll contribute to a mission-driven platform that enables reliable AI systems at scale.

Responsibilities

• Own large areas of platform end to end, from design to production deployment
• Develop knowledge representation systems including ontologies and knowledge graphs
• Design and implement RAG pipelines (chunking, embeddings, indexing, retrieval, reranking)
• Build integrations between ML components and diverse enterprise data sources, vector databases, APIs, and services
• Develop context retrieval systems balancing recall, precision, latency, and cost
• Create evaluation frameworks, datasets, and metrics for retrieval quality and end-to-end agent performance
• Build reliable backend services and data pipelines to support ML and LLM components in production
• Deliver experiments and new capabilities quickly with high quality and customer feedback loops
• Collaborate across product, ML, and infrastructure teams to shape platform direction

Key requirements

• 5+ years of experience building and deploying ML/AI systems in production
• Deep hands-on understanding of retrieval systems, RAG, embeddings, vector indexing, and knowledge representation
• Experience with knowledge representation, semantic search, or agentic systems
• Proficiency in Python with production-quality, testable, maintainable code
• Experience shipping products at high-growth startups
• Ability to operate in ambiguous problem spaces balancing research with pragmatic constraints
• Strong communication skills and comfort in customer-facing or cross-functional environments

• strong communication skills
• cross-functional collaboration
• customer-facing comfort
• retrieval systems
• RAG pipelines
• embeddings

…