Longitude Labs - Member of the Technical Staff

deCircle; New York (NY); On-site; USD 250,000 - 350,000; Full time

Job summary

Longitude Labs is seeking a leading AI research hire to design benchmarks, evals and dataset taxonomies that define high-quality AI data. You will perform independent research and publish to establish thought leadership in AI benchmarks.

In-office in NYC or SF (5 days, flex 4) with visa sponsorship available; base salary 250k–350k plus competitive equity. PhD in ML/CS/CV or quantitative science strongly preferred and top-venue publications valued.

Qualifications

• PhD in ML/CS/CV or quantitative science (strongly preferred)
• Top-venue publications (ICML, NeurIPS), ideally top author with cited work
• Deep LLM, benchmark design, eval methodology and data quality expertise
• NYC or SF, in-office 5 days (flex 4)

Responsibilities

• Design and own benchmarks, evals and dataset taxonomies setting the standard for AI data quality
• Independent research and publishing to establish thought leadership in AI benchmarks
• Identify where frontier models fail and build datasets to close gaps
• Partner with domain experts and deployment leads on what high-quality data looks like
• Drive the research agenda with autonomy

Skills

Deep LLM

Benchmark design

Evaluation methodologies

Data quality

Python

PyTorch

LLMs

Data Pipelines

Education

PhD in ML/CS/CV or quantitative science

Tools

Python

PyTorch

LLMs

Evaluation Frameworks

Data Pipelines

Job description

Longitude Labs: stealth data infrastructure for leading AI labs (30 people, $30M raised; founders from a top frontier lab and an AI-infra decacorn). One of the first dedicated research hires.

Comp: $250K-$350K base (can go higher for exceptional research credentials) + competitive equity. Location: In-office 5 days (flexible to 4), NYC or SF. Visa: Visa transfers and sponsorship available.

Responsibilities

• Design and own benchmarks, evals and dataset taxonomies setting the standard for AI data quality

• Independent research and publishing to establish thought leadership in AI benchmarks

• Identify where frontier models fail and build datasets to close gaps

• Partner with domain experts and deployment leads (life sciences, finance) on what high-quality data looks like

• Drive the research agenda with autonomy

Tech stack: Python, PyTorch, LLMs, Evaluation Frameworks, Data Pipelines

Source: https://www.paraform.com/share/longitude-labs/cmrdwtlj300000djmmm0hmpjy

• PhD in ML/CS/CV or quantitative science (strongly preferred)

• Top-venue publications (ICML, NeurIPS), ideally top author with cited work

• Deep LLM, benchmark design, eval methodology and data quality expertise

• NYC or SF, in-office 5 days (flex 4)