Quantitative Research Scientist

OpenHouse.ai; Calgary; On-site; CAD 80,000 - 135,000; Full time

Benefits offered by this job

Stock options

Competitive compensation

Job summary

OpenHouse.ai in Calgary, Canada is seeking a Quantitative Research Scientist to turn messy real-world systems into tangible problems. You will formulate hypotheses, select methods, and determine if models improve decisions.

You will work on forecasting, causal inference, operations research, simulation, and optimization to solve pricing, inventory, and construction- decision problems. You will own problems from framing through deployment and learning from outcomes.

Qualifications

• A track record of exceptional quantitative work.
• Strong quantitative foundations in probability, statistics, and mathematical modeling.
• Strong modelling judgement and the ability to identify what matters in a problem.
• Strong programming ability to turn ideas into experiments, simulations, and prototypes.
• Research judgment to formulate hypotheses and distinguish evidence from speculation.
• Clear technical reasoning and explicit assumptions with transparent uncertainty.

Responsibilities

• Frame hard problems by turning business and operational data into quantitative formulations.
• Choose appropriate modeling approaches (forecasting, causal inference, simulation, optimization).
• Design convincing evaluations with baselines, backtests, and sensitivity analyses.
• Extract signal from imperfect, sparse, delayed data and understand data generation processes.
• Close the research loop by prototyping and integrating into real decision workflows.
• Bring outside ideas from statistics, ML, economics, and OR and assess their transferability.
• Try to disprove assumptions by running sensitivity analyses and exploring competing explanations.

Job description

OpenHouse.ai is building the intelligence backbone for homebuilding. Our quantitative systems are deployed across more than 30 markets in the U.S. and Canada, supporting hundreds of new construction communities and helping builders make better decisions about pricing, inventory, construction, and capital.

Housing gives us a rare research environment: consequential decisions are made repeatedly under uncertainty, markets evolve continuously, interventions affect future observations, and the data are sparse, heterogeneous, and deeply connected to how the underlying business operates.

About The Role

As a Quantitative Research Scientist, your job is to turn messy real-world systems into tangible quantitative problems. You will formulate hypotheses, identify the structure that matters, choose or develop appropriate methods, and determine whether the resulting models actually improve decisions.

The work draws on forecasting, causal inference, operations research, simulation, optimization, and decision‑making under uncertainty. Many of the problems we encounter have no established solution, and solving them often requires questioning the industry's conventional framing before developing a new one.

Modern AI tools dramatically increase the leverage available to a strong quantitative researcher: accelerating literature review, hypothesis generation, experimentation, implementation, and analysis. We expect researchers here to aggressively embrace that leverage.

Researchers own problems end‑to‑end, from framing and research through experimentation, implementation, deployment, evaluation, and learning from real‑world outcomes.

What You'll Work On

Our research spans problems such as:

• Market & Pricing Science: Quantify demand patterns, price elasticity, market dynamics, and causal impacts in environments characterized by sparse interventions, delayed feedback loops, and shifting market conditions.
• Decision Science: Build models that help builders reason about tradeoffs between margin, sales velocity, inventory exposure, timing, and uncertainty.
• Construction Operations: Use forecasting, queueing theory, and operations research to predict cycle time and uncover latent operational constraints.
• Optimization: Explore how builders should allocate scarce resources and sequence decisions across communities, inventory, construction capacity, and capital.

Many of these problems do not have established solutions. In other cases, the industry's conventional framing embeds assumptions that deserve to be tested. A core part of the researcher's job is deciding whether the question itself is the right one.

What You'll Do

• Frame hard problems. Turn ambiguous business, financial, and operational systems into quantitative formulations. Identify the key state variables, constraints, assumptions, and mechanisms.
• Choose the right level of model complexity. Develop forecasting, causal, simulation, optimization, or statistical approaches based on the structure of the problem, and know when a simpler model is more useful than a sophisticated one.
• Design convincing evaluations. Build falsifiable tests, baselines, backtests, counterfactual analyses, simulation studies, or field experiments. Actively search for alternative explanations and evidence that could disprove the model.
• Extract signal from imperfect observations. Work with sparse, delayed, heterogeneous, and selectively generated data. Understand the nature of operational processes and how data are captured before deciding what can legitimately be inferred from them.
• Close the research loop. Turn promising ideas into working prototypes, collaborate with engineers to put them into real decision workflows, measure what happens, and use those outcomes to improve the underlying model.
• Bring outside ideas into the problem. Follow relevant work across statistics, machine learning, economics, operations research, quantitative finance, and adjacent disciplines, then determine which ideas actually transfer to the systems we are studying.
• Try to prove yourself wrong. Identify hidden assumptions, construct competing explanations, run sensitivity analyses, and understand where a model is likely to fail before relying on it.

You will own quantitative problems from first principles through real‑world outcomes.

What You'll Bring

We care more about demonstrated quantitative ability than a particular credential or career path. Strong candidates may come from applied mathematics, statistics, physics, operations research, economics, quantitative finance, machine learning, engineering, or other technically rigorous fields.

What matters most is evidence that you can independently solve difficult quantitative problems.

You should bring:

• A track record of exceptional quantitative work. You have solved technically difficult problems where the answer, formulation, or appropriate methodology was not obvious in advance. Evidence may come from research, publications, production systems, open‑source work, competitions, graduate work, or demanding industry projects.
• Strong quantitative foundations. You have deep foundations in probability, statistics, mathematical modelling, and quantitative reasoning, with the ability to work in areas such as optimization and decision‑making under uncertainty as the problem requires.
• Strong modelling judgment. You can identify which aspects of a problem matter, which can safely be ignored, and the simplest approach that preserves the important structure.
• Strong programming ability. You can turn mathematical ideas into reliable experiments, simulations, prototypes, and working implementations.
• Research judgment. You can enter an ambiguous problem, formulate useful hypotheses, distinguish important questions from interesting distractions, and determine what evidence would meaningfully change your view.
• Clear technical reasoning. You make assumptions explicit, explain why you chose an approach, communicate uncertainty, and distinguish what the evidence supports from what remains speculative.
• Intellectual range and learning velocity. When a problem requires unfamiliar methods, you can learn the relevant ideas quickly, evaluate them critically, and put the useful ones to work.

Your ability to deeply understand a problem and quickly learn the right tools is more important than experience with specific frameworks.

You enjoy difficult problems where the answer is not obvious.

You will likely do well at OpenHouse.ai if:

• You like bringing structure to messy problems.
• You are curious about why something works, not simply whether it works.
• You would rather discover that an idea is wrong quickly than spend months defending it.
• You prefer the simplest explanation or model that captures what matters.
• You want your work exposed to real decisions and real consequences.
• You are comfortable owning a problem before the path to the answer is clear.

OpenHouse.ai is still small enough that strong researchers can materially influence what problems we pursue, how we approach them, and what gets built. That also means priorities evolve and researchers are expected to exercise judgment rather than wait for perfectly specified problems.

How We Work

Three principles shape how we work at OpenHouse.ai:

• Uncover the Unknown: We treat assumptions as hypotheses, design tests that can prove us wrong, and follow the evidence, even when it means reframing the problem.
• Own the Outcome: We expect end‑to‑end ownership: understand the problem, develop the idea, test it, build it, deploy it, and learn from the results.
• Cultivate Idea Meritocracy: Ideas stand on the strength of their reasoning and evidence. We challenge each other thoughtfully and change our minds when the evidence changes.

What We Offer

Our compensation and benefits include:

• $80,000-$135,000annual salary, depending on experience, skills, and level
• Stock options, giving you the opportunity to share in the long‑term value you help create