About the role
Can you shape a smarter, more scientific way of discovering talent?
At Unstop, we're looking for a People Scientist at Unstop, who will own the science behind our assessments, from competency model to validated, client-ready test. You will work with psychologists and engineers to decide what our products measure, how they measure it, and how we prove it works. You will create new assessments and methods, find and fix gaps in existing ones, and bring generative AI and AI-based scoring into how we build and score tests, with the rigour of a psychometrician. You will also represent our science to clients, and help win new business.
What you'll own
- Build, validate, and maintain competency models through job analysis, and set the assessment strategy for each client need, including which constructs and methods to use.
- Design, develop, and validate new assessments, including personality measures and competency-based simulations such as caselets, situational judgement tests, in-basket exercises, role plays, and assessment centre exercises.
- Use generative AI to scale item and content development, and design and validate AI-based scoring of open-ended, video, and conversational responses against expert ratings, with checks for bias and drift.
- Lead validation research and advanced analytics (e.g., CTT, IRT, SEM, machine learning), and audit existing assessments to find and fix gaps such as weak items, low reliability, or adverse impact.
- Partner with product and engineering teams to turn the science into scoring logic, item banks, and reports, and with sales teams to write proposals and business pitches.
- Present findings to clients, write technical and general content (e.g., validation reports, whitepapers, blogs), and mentor People Science Associates.
Requirements
- Master's degree in Industrial/Organisational Psychology, Psychometrics, or a related field.
- 3 years of experience in assessment design, people science, or talent consulting.
- Experience developing and validating psychometric tests, including personality assessments and competency-based simulations (e.g., caselets, situational judgement tests, in-basket exercises).
- Experience with classical test theory, item response theory, and multivariate statistics (e.g., EFA/CFA, SEM, cluster analysis) using R, Python, or SPSS.
- Experience using generative AI or AI-based scoring methods in assessment development.
Good to have
- PhD in Industrial/Organisational Psychology, Psychometrics, or a related field.
- Experience designing and validating AI-based scoring of open-ended responses (e.g., NLP, LLM-based, or machine-learning scoring), including agreement and fairness analysis against expert ratings.
- Experience building LLM-assisted item generation workflows with human review and psychometric quality checks.
- Experience with machine-learning models, computerised adaptive testing, or large-scale item banks.
- Experience writing client proposals and business pitches, and presenting to senior HR and business stakeholders.
- Knowledge of cognitive psychology, professional standards (e.g., SIOP Principles, APA/AERA/NCME Standards, ISO 10667), and rules on AI in hiring and data privacy (e.g., India's DPDP Act).
- Experience with training development or instructional design.
- Excellent technical and business writing skills.