Small teams. Every paper has an author you can write to.
We are deliberately small. Every published result names the team that did the work, and that team will answer a question about the method, rather than a press address that routes to nobody.
Who works on what.
Mohit Bhardwaj
Head of Analytics, December 2018 to December 2022 · Predictive modelling · Causal inference · Statistical reporting & privacyLed the analytics group from December 2018 to December 2022: discrete-time hazard models for attrition, correcting models after the 2020 move online, doubly robust programme evaluation across colleges, shrinkage for result analyses, and disclosure control for published returns. Author of the five reports listed below.
- Small cells, big leaks: reconstruction attacks on institutional returns, and what differential privacy costs · EPR-2022-02
- Small sections, loud numbers: empirical Bayes shrinkage for course-level result analysis · EPR-2022-01
- Doubly robust estimates of programme effects when uptake clusters by college · EPR-2021-01
- When the term moved online: detecting and correcting dataset shift in early-warning models · EPR-2020-01
- Time to departure: discrete-time hazard models for student attrition · EPR-2019-01
EduPlatter Modelling team
Research group · Predictive modellingBuilds and validates the institution-specific persistence and completion models, and the gates that stop a weak model reaching an advisor.
EduPlatter Efficacy team
Research group · Causal inferenceRuns the termly matched-comparison studies that tell an institution whether an initiative changed anything.
EduPlatter Responsible Analytics team
Research group · FairnessOwns subgroup fairness reporting, the deployment gate it feeds, and the rules for what may be published about small groups.
EduPlatter Curriculum Analytics team
Research group · Curriculum analyticsWorks on course sequencing, capacity and demand forecasting: the structural reasons students finish late.
EduPlatter Data Engineering team
Research group · Data qualityIntegrates institutional systems and measures what each source is actually worth to a model.
- Data quality is the binding constraint, not model choice · EPR-2025-02
EduPlatter Student Voice team
Research group · Natural languageAnalyses open-text feedback in the multilingual, code-mixed reality of Indian classrooms, in aggregate only.
People who tell us when we’re wrong.
Three external advisors review methodology before publication. They are paid for their time, they hold no equity, and they are free to disagree with us publicly. Two have.
Prof. Meera Iyer
Educational measurement, Indian Institute of Education (visiting)
Reviews study design before publication and has twice told us a result would not survive peer review.
Dr. James Okonkwo
Learning analytics, independent
Advises on the fairness auditing programme and the limits of what our gap definition can establish.
Anjali Varma
Data protection and student rights, independent counsel
Advises on what our processor commitments should mean in practice, including where they cost us revenue.
Research roles here ship.
Every area lead also owns the corresponding production system. The measurement and the implementation are the same job, which is either the appeal or the objection depending on what you want from a research post.
You publish what you find
No internal review can suppress a result for being commercially inconvenient. Two of our eleven papers contain findings that are bad for us.
Prior work stays yours
We do not assign anything you built before you arrived.
Real institutional data, real constraints
You will work with live student records under processor obligations, which is more interesting and more restrictive than a public dataset.