Member of Engineering (AI)

TL;DR

Member of Engineering (AI): Building and optimizing high-quality pretraining datasets and synthetic data generation pipelines for LLMs with an accent on scalability, diversity, and model capability mapping. Focus on designing trillion-scale data pipelines, conducting quantitative ablation experiments, and leveraging large GPU clusters to improve coding agent performance.

Location: Remote (Must be based in the US)

Company

hirify.global is an AI company building frontier models and agentic systems to accelerate software development and reach AGI.

What you will do

• Design and implement complex pipelines for large-scale synthetic data generation while optimizing available resources.
• Improve the quality of pretraining datasets by leveraging research, intuition, and training experiments.
• Collaborate with Pretraining, Post-training, Evals, and Product teams to map data needs to missing model capabilities.
• Lead original research initiatives through time-bounded experiments and deploy technical solutions into production.
• Measure and refine dataset quality through quantitative data ablation experiments.

Requirements

• Strong background in machine learning and engineering.
• Deep experience with LLMs, including scaling laws, post-training techniques, and training reasoning models.
• Proven track record of building trillion-scale pretraining datasets, including data curation, deduplication, and mixing.
• Expertise in Python and strong prompt engineering skills.
• Experience working with large-scale GPU clusters and distributed data pipelines.
• Must be based in the US.

Nice to have

• Author of scientific papers on applied deep learning, LLMs, or source code generation.

Culture & Benefits

• Fully remote work and flexible hours.
• Generous leave policy with 37 days of vacation and holidays per year.
• Health insurance allowance for employees and their dependents.
• Company-provided equipment and allowances for home office and continuous learning.
• Regular team gatherings, including monthly 3-day meetups in Paris.

Hiring process

• Introductory call with a Founding Engineer.
• Technical interviews with Members of Engineering.
• Team fit call with the People team.
• Final interview with a Founding Engineer.