Build synthetic data systems that turn specialized workflows into realistic, scalable training tasks for AI agents. You will work with an engineering team to develop task generation, validation, and evaluation methods that help improve model capabilities.
Build pipelines that transform domain-specific workflows into structured and challenging synthetic training tasks.
Collaborate with subject-matter experts to create tasks across professional and technical domains.
Develop methods for generating diverse, realistic, and learnable tasks.
Create tooling to mutate, validate, and improve synthetic tasks.
Analyze model and agent performance to identify what tasks teach and where they fail.
Develop metrics for task diversity, realism, learnability, and quality.
At least 2 years of experience in software engineering, machine learning engineering, or AI research, building and delivering data pipelines, ML infrastructure, or synthetic data systems.
Technical education or background, with hands-on experience developing end-to-end synthetic data pipelines for AI or machine learning applications.
Proficiency in Python, Linux development, and containerization tools such as Docker.
Experience evaluating synthetic data quality, including diversity, realism, learnability, and limitations.
Experience building evaluation frameworks, benchmarks, or testing environments for AI agents or large language models.
Ability to independently deliver technical projects with minimal predefined requirements, and to identify subtle data inconsistencies and edge cases.
Strong reasoning about task design, scoring, and failure modes; familiarity with reinforcement learning, agent workflows, or LLM post-training is useful.
Clear communication skills for collaboration across time zones.
Base salary: $100,000 to $170,000 USD annually. Visa sponsorship is available.
On-site in Singapore, Singapore.