Platform Engineer for Frontier Model Pretraining

Cursor; United States; On-site; USD 120,000 - 170,000; Full time

Job summary

Triwill Group is building data systems to support frontier coding models. You will join one of the core teams—Data Quality, Data Platform, or Crawling—to transform raw web data into training-ready datasets and to own scalable pipelines end-to-end.

The role emphasizes high-throughput processing, traceability, and close collaboration with researchers to drive model performance. The team values ownership, exploration, and practical engineering excellence.

Qualifications

• Experience with large-scale distributed systems and data pipelines.
• Ability to ship end-to-end data software with high ownership.
• Familiarity with data quality and training data workflows.

Responsibilities

• Build and own high-throughput data pipelines with end-to-end traceability.
• Train and deploy models that classify, rank, clean, and filter data at extreme throughput.
• Design scaling experiments to improve data quality and dataset usefulness.
• Collaborate with Data Acquisition and training teams to improve loss and evals.
• Write performance-critical code and design rigorous experiments.

Skills

Distributed systems

Data pipelines

Web crawling

Telemetry pipelines

Job description

Triwill Group is building data systems to support frontier coding models. You will join one of the core teams—Data Quality, Data Platform, or Crawling—to transform raw web data into training-ready datasets and to own scalable pipelines end-to-end.

The role emphasizes high-throughput processing, traceability, and close collaboration with researchers to drive model performance. The team values ownership, exploration, and practical engineering excellence.