Senior Lead ML Encoder for Customer-360 Embeddings

Aquent; South San Francisco (CA); On-site; USD 125,000 - 139,000; Full time

Benefits offered by this job

Health insurance

Vision insurance

Dental insurance

Paid sick leave

401(k) with matching

Training opportunities

Job summary

Aquent partners with a major Biotechnology company to find a Senior Lead ML Encoder for a temporary opportunity. You will lead the design and execution of a first-of-its-kind customer-360 representation model—training a dense vector embedding on longitudinal transaction and interaction history to power GenAI and analytics products across the enterprise.

You will collaborate with data scientists and product teams, implement robust evaluation and production pipelines in PyTorch or JAX, and present

Qualifications

• H hands-on experience designing pretraining objectives and training encoder/embedding models from scratch.
• Deep expertise in representation learning techniques (contrastive learning, temporal sequences, Transformers, GNNs).
• Proven track record modeling large, sparse longitudinal event data.
• Experience building inductive representations generalizing with minimal history.
• Strong Python engineering with PyTorch or JAX and distributed data frameworks.

Responsibilities

• Design pretraining objectives and build representations from longitudinal data.
• Develop, train, and evaluate encoder models using PyTorch or JAX for scalable distributed processing.
• Establish robust evaluation frameworks with time-based splits and leakage detection.
• Write production-grade Python code to move models from research to deployment.
• Assess embedding signal by calibration, drift, and subgroup performance.
• Present findings and recommendations to senior stakeholders.

Skills

Pretraining objectives design

Representation learning

PyTorch

JAX

Python

Longitudinal data modeling

Model evaluation

Tools

SQL

Distributed training

Cloud training at scale

Job description

Aquent partners with a major Biotechnology company to find a Senior Lead ML Encoder for a temporary opportunity. You will lead the design and execution of a first-of-its-kind customer-360 representation model—training a dense vector embedding on longitudinal transaction and interaction history to power GenAI and analytics products across the enterprise.

You will collaborate with data scientists and product teams, implement robust evaluation and production pipelines in PyTorch or JAX, and present