Tavern Research is building models to understand and influence how people form opinions online, addressing the challenge of dynamic systems involving human behavior and the spread of information, narratives, and public opinion, especially in the context of political campaigns and the future of democracy.
Requirements
- Advanced proficiency in Python and modern ML/AI libraries, with ability to write production-quality code.
- Extensive experience working with messy, large-scale datasets.
- Depth in advanced techniques such as causal inference, topic modeling, network analysis, embeddings, optimization, or agentic systems, with strong grounding in experiment design and statistical evaluation.
- Demonstrated ability to evaluate models rigorously and iterate quickly.
- Collaborative mindset and comfort with modern engineering workflows (version control, testing, code review).
- Prior experience mentoring or supporting junior modelers, and exposure to production ML pipelines.
- 4+ years of relevant experience who can independently lead complex modeling projects.
Responsibilities
- Lead the design and analysis of experiments using methods like randomized trials, difference-in-differences, and causal inference.
- Build, maintain, and evaluate advanced statistical and machine learning models that generate political and behavioral insights.
- Engineer features and build scalable pipelines from messy, real-world datasets.
- Write clean, production-ready Python code and documentation.
- Drive rigorous model evaluation, diagnostics, and iteration to ensure reliability.
- Collaborate with researchers and engineers to translate open-ended questions into solvable modeling approaches.
- Contribute to the team’s technical direction by introducing new methods, tools, or frameworks.
Other
- This role is designed for experienced modelers with 4+ years of relevant experience who can independently lead complex modeling projects.
- Proven track record of leading substantive modeling projects that delivered real-world impact.
- Strong communication skills with ability to explain complex concepts clearly to technical and non-technical audiences.
- We value in-person work and expect employees to work regularly from our Chicago office.
- Provide peer support and mentorship to junior modelers, helping to grow team capacity.
