Multimodal ML Engineer — Production Systems

Jack; San Francisco (CA); On-site; USD 150,000 - 350,000; Full time

Job summary

Sieve in San Francisco is seeking a Machine Learning Engineer to own end-to-end ML problems, from understanding customer needs to building production pipelines and evaluating models. You'll work on multimodal data, data curation, model deployment, and evaluation systems, collaborating with frontier AI teams to ship high-quality datasets and robust ML solutions.

This role suits engineers who thrive on ambiguity, enjoy prototyping with new models, and can translate complex requirements into

Qualifications

• 2+ years of relevant experience.
• Experience building and shipping production machine learning systems.
• Experience training, fine-tuning, or deploying modern deep learning models.
• Hands-on experience with PyTorch.
• Experience working with modern foundation models.
• Strong understanding of model evaluation, dataset quality, precision and recall tradeoffs, and edge cases.
• Ability to own technical projects from problem definition through deployment.
• Strong communication skills and comfort working with customers and cross-functional teams.
• Able to work onsite at Sieve’s San Francisco office 5 days per week.

Responsibilities

• Build and improve production machine learning systems in Python
• Train, fine-tune, and deploy modern deep learning and foundation models
• Develop evaluation and QA pipelines using models such as GPT, Claude, Gemini, and open-source models
• Design systems for filtering, ranking, retrieval, labeling, and dataset curation
• Build benchmarks and evaluation frameworks for model and dataset quality
• Analyze precision, recall, edge cases, and failure modes to improve system performance
• Develop scalable ML pipelines spanning preprocessing, inference, post-processing, and quality validation
• Work directly with customers and cross-functional teams to translate complex requirements into production systems
• Rapidly prototype with new models, tools, and APIs

Skills

Python programming

Production ML systems

Foundation models

Model evaluation

Customer collaboration

Problem ownership

Deployment of ML pipelines

Fast-moving team communication

Tools

GPT/Claude/Gemini

Open-source models

Job description

Sieve in San Francisco is seeking a Machine Learning Engineer to own end-to-end ML problems, from understanding customer needs to building production pipelines and evaluating models. You'll work on multimodal data, data curation, model deployment, and evaluation systems, collaborating with frontier AI teams to ship high-quality datasets and robust ML solutions.

This role suits engineers who thrive on ambiguity, enjoy prototyping with new models, and can translate complex requirements into