Forecasting Research Scientist

• Frame forecasting questions with a stated horizon and a rule for how each one resolves.
• Design agents that read market data, news and filings and state a forecast with a probability and a rationale.
• Record every forecast with its time, inputs, model version and rationale before the outcome is known.
• Score forecasts out of sample with proper scoring rules, against simple baselines and runs on randomly chosen dates.
• Check calibration by comparing stated probabilities with observed frequencies.
• Build backtests that replay only what was known at each point in time.
• Correct significance thresholds when many hypotheses are tested at once, and report misses beside hits.
• Run event studies of how prices respond to announcements and filings, against random baseline windows.
• Write up methods and results for clients and for research reports.

• An MS or PhD in statistics, economics, computer science or another quantitative field, or equivalent experience.
• 3+ years in forecasting, time-series or econometric research.
• Experience with proper scoring rules, calibration and out-of-sample testing.
• Python for data analysis and modeling.
• Clear writing about methods and uncertainty.

• Experience building research tools with LLM agents.
• Event study methods.
• Bayesian or hierarchical models.
• Published forecasting research or a forecasting tournament record.

$117,000 to $157,000 a year

Full-time W-2 employment with health insurance, a retirement plan and paid leave.

No application deadline. We accept applications for this role on an ongoing basis.

Remote, United States