Research, data, and AI questions worth working through carefully.
I work with researchers and teams on difficult questions involving study design, data, language, and applied AI. The useful part is often deciding what to measure, which method fits, and what the result can actually support.
What I bring
A clear question, a defensible method, and a useful handoff.
We agree on the question, scope, responsibilities, and final deliverable before the work begins.
The plan includes the timeline, review points, and what I need from you.
Ways I can contribute
Research design, data analysis, and applied AI.
Each collaboration starts with a defined question and ends with a result you can inspect, use, or build on.
Research design
Clarify the question before choosing the method.
I work with researchers on study design, measurement, analysis plans, and the tradeoffs behind a defensible answer.
- Research questions and study design
- Survey, experiment, and measurement planning
- Analysis strategy and robustness checks
- Clear documentation of decisions
Data and computational methods
Make difficult data easier to reason with.
I help structure, analyze, and communicate complex behavioral or text data using R, Python, statistics, and machine learning.
- Data cleaning and reproducible analysis
- Statistical modeling in R or Python
- Text analysis and classification
- Visualization and interpretation
Applied AI
Decide where AI belongs in the workflow.
I help teams think through retrieval, evaluation, professional review, and what a useful prototype should actually do.
- Workflow and information mapping
- Retrieval and document design
- Evaluation plans and failure modes
- Prototype and product critique
How we work
From question to handoff.
A first conversation is enough to see whether my background fits the problem.
- 01
Describe the question
Share the decision you need to make, the data or materials you have, and what has already been tried.
- 02
Choose the method
I outline an approach, explain its tradeoffs, and make the scope explicit.
- 03
Work through the analysis
I build or review the agreed analysis, workflow, or prototype and document the important decisions.
- 04
Review and hand off
We review the result together, resolve open questions, and leave you with clear documentation.
Background
Work directly with me.
I’m Jonathan Doriscar, a cognitive scientist and computational behavioral scientist at Northwestern. My training spans psychology, statistics, data science, and machine learning.
As a student consultant with Northwestern IT Research Computing, I work with researchers on applied statistics, R and Python, visualization, and research computing.
Academic affiliations and fellowships are background information only. They do not imply that Northwestern University or the National Science Foundation endorses this work.
Usually a good fit
- You have an important question but are unsure which design or method fits.
- Your behavioral, survey, or text data needs a defensible analysis.
- You want to evaluate an AI workflow before investing in a larger build.
- You value clear reasoning and direct collaboration.
Have a question?
Tell me what you’re working on.
A few sentences about the question, data, or workflow are enough to begin.