Senior ML Platform Engineer: Semantic Search & Multimodal

stackav; United States; Remote; USD 150,000 - 210,000; Full time

Job summary

stackav is seeking an experienced engineering leader to shape and build semantic search and data discovery systems for text, images, and video. You will drive inference services, vector database integrations, and end-to-end data mining capabilities.

You will explore language and vision-language models, develop real-time query services, and design batch or streaming pipelines for large multimodal datasets. Collaboration with partner teams to set technical direction is essential.

Qualifications

• Five+ years of experience building ML platforms and applications.
• Experience delivering scalable, reliable infrastructure in fast-moving environments.
• Hands-on experience with semantic search, multimodal embeddings, and vector databases.
• Experience with multimodal indexing, training infrastructure, model optimization, or large-scale data processing.
• Ability to explain architectural tradeoffs and align stakeholders on design decisions.

Responsibilities

• Shape and build semantic search and data discovery systems for text, images, and video.
• Explore language and vision-language models and retrieval techniques to improve event discovery and curation workflows.
• Develop real-time query services and batch or streaming pipelines for large multimodal datasets.
• Connect onboard software with offboard machine learning and data infrastructure to support end-to-end data mining.
• Identify infrastructure bottlenecks, improve system performance, and write efficient Python or C++.
• Set technical direction with partner teams and deliver solutions across organizational boundaries.

Skills

Semantic search

Multimodal embeddings

Vector databases

ML infrastructure

Model optimization

Data processing

Architectural tradeoffs

Tools

Ray

Spark

Lance

Iceberg

Job description

stackav is seeking an experienced engineering leader to shape and build semantic search and data discovery systems for text, images, and video. You will drive inference services, vector database integrations, and end-to-end data mining capabilities.

You will explore language and vision-language models, develop real-time query services, and design batch or streaming pipelines for large multimodal datasets. Collaboration with partner teams to set technical direction is essential.