Foundational Data Engineer

Stack

Responsibilities

• ▹Make Normal's models better at hardware design, verification, and EDA workflows by any means possible
• ▹Own the data flywheel from the company's own agent runs: rejection sampling, distillation, mining eval-passing trajectories
• ▹Identify, evaluate, and acquire datasets relevant to hardware design, verification, and EDA workflows
• ▹Partner with verification engineers to define quality rubrics and curate golden reference examples
• ▹Operate data ingestion pipelines, monitor for quality regressions and coverage gaps
• ▹Negotiate customer-data access on-prem and federated, handling redaction and IP constraints
• ▹Build the data team as it scales, across synthetic data, verification SME curation, and data infrastructure

Requirements

• ▹Have built or used a data flywheel: model outputs curated into the next training round
• ▹Approach data acquisition as an engineering problem: systematic, measurable, outcome-driven
• ▹Shipped a synthetic-data or training-data pipeline that produced measurable downstream model improvement
• ▹Can evaluate data quality independently, spotting noise, bias, and gaps
• ▹Comfortable working across multiple technical roles, synthesizing feedback from domain experts, ML engineers, and pipeline engineers
• ▹Organized and documentation-minded: tracks provenance, ownership, and lineage as a matter of habit

Nice to have

• ▹Experience acquiring data from a variety of paid and unpaid sources, managing vendor relationships
• ▹Familiarity with SystemVerilog, Verilog, and UVM
• ▹Background in code-model or agent training-data pipelines (e.g. SWE-bench-style data, code-model post-training)
• ▹Experience with automated data collection, web scraping, or corpus curation at scale
• ▹Prior work in a startup or fast-moving research environment where the data strategy was still being defined

Soft skills

About the company

Normal Computing builds silicon that turns thermal noise from an obstacle into a computational resource — stochastic, in-memory, asynchronous architectures that deliver 10-100x more AI inference per dollar and per watt. The company co-designs the full stack, from AI-native EDA systems used by the world's largest semiconductor companies to the advanced ASICs they make possible. Backed by $85M+ from leading deep-tech investors, the team works across New York, Silicon Valley, London, Copenhagen, and Seoul.