Applied Research & Data Scientist (Learning Analytics & Multimodal AI)

Type: Full-time / Onsite (or Hybrid 1 day WFH)

Level: Senior

About the client

Our client is a fast growing EdTech company with a mission to help individuals optimize their learning through AI. They leverage cutting-edge technology to provide personalised learning experiences, making education more efficient and effective. 

They are a high-growth tech startup on a mission to make education accessible to 1 billion children.

The Role

They are looking for an Applied Research & Data Scientist who loves exploring how people learn. You'll lead research and modeling efforts across handwriting and voice datasets, blending regression analysis, causal inference, and machine learning to test hypotheses, uncover learning patterns, and build predictive models.

This role sits at the intersection of data science and applied AI research where statistical rigor meets creative experimentation. You'll collaborate with product and pedagogy teams to turn quantitative insights into real-world educational impact.

You will be pioneering how AI understands human learning. By analysing handwriting and voice data, they uncover how students think, feel, and learn identifying behavioral signals like stroke speed, hesitation, tone, and timing that correlate with learning outcomes.

The goal is to transform messy, multimodal data into meaningful insights that guide each students next best learning step.

Key Responsibilities

Analytical Research & Modelling

• Conduct exploratory data analysis (EDA) on handwriting and voice datasets to identify behavioral patterns and anomalies.
• Build and interpret regression models linear, logistic, mixed effects, LASSO,
• Ridge to isolate key factors influencing performance, engagement, or stress.
• Use multivariate and non-linear regression to examine interdependent behavioral
• relationships (e.g., writing acceleration vs. tone modulation).
• Apply causal inference frameworks backdoor criterion, DAGs, propensity scoring, mediation analysis to uncover genuine cause-effect linkages.
• Perform feature engineering from raw handwriting and audio data (e.g., hesitation index, cognitive delay markers, pitch variability).
• Employ dimensionality reduction (PCA, UMAP, t-SNE) and unsupervised learning (clustering, mixture models) to discover latent learning traits.
• Quantify uncertainty, confidence intervals, and perform model diagnostics (VIF, residual analysis, cross-validation).

Machine Learning & AI Integration

• Develop predictive models to forecast engagement, confidence, or completion speed.
• Experiment with speech emotion recognition, sequence models, or multimodal fusion networks that integrate handwriting + audio.
• Collaborate with engineers to prototype LLM-driven insight layers that summarize behavioral findings or explain patterns.
• Use NLP and embedding models to analyse transcribed speech or open-ended answers for affective or cognitive insights.

Research & Hypothesis Testing

• Design and conduct experiments, quasi-experiments, or A/B tests to validate hypotheses and interventions.
• Use statistical hypothesis testing (ANOVA, chi-square, t-tests, permutation testing) to verify observed trends.
• Apply causal reasoning to determine which variables most strongly influence learning efficiency.
• Build reproducible research pipelines (Jupyter, MLflow, W&B) with versioncontrolled analysis and documentation.

Data Infrastructure & Visualization

• Work with engineers to maintain clean, labeled, and reliable multimodal data pipelines.
• Optimize ETL workflows for handwriting and voice signals, ensuring high-quality data ingestion.
• Create interactive dashboards or visualizations (Streamlit, Plotly, Tableau) to communicate insights intuitively.
• Document datasets, model assumptions, and findings in clear technical and narrative formats.

Collaboration & Communication

• Partner with educators and product managers to interpret results in a learning context.
• Translate analytical outputs into actionable insights for curriculum and product design.
• Present complex analyses in simple, visual, and narrative forms for non-technical stakeholders.
• Support leadership with metrics that guide pedagogy strategy, AI model improvements, and product direction.

Qualifications

Must-Have

• Bachelors or Masters degree in Data Science, Statistics, Machine Learning, AI, Cognitive Science, or related field.
• Proven experience in regression modeling (linear, logistic, hierarchical, regularized) and causal inference.
• Proficiency in Python (pandas, NumPy, scikit-learn, statsmodels, PyTorch/TensorFlow, causalml, DoWhy).
• Strong grasp of statistical hypothesis testing, experimental design, and model diagnostics.
• Experience in feature engineering, dimensionality reduction, and unsupervised learning.
• Excellent analytical reasoning, hypothesis formulation, and data storytelling skills.
• High level of proficiency in English.

Good-to-Have

• Exposure to LLMs, prompt engineering, and multimodal representation learning.
• Familiarity with Bayesian modeling, causal ML, or hierarchical models.
• Background in educational data mining, learning analytics, or human performance modeling.
• Experience integrating data models into AI-driven insights dashboards or feedback loops.
• Experience with speech signal processing (librosa, OpenSMILE) or handwriting trajectory analysis (CNN/RNN-based stroke models).

Why Join them

• Opportunity to build a global movement and shape the frontier of how AI understands human learning.
• Opportunity to work with an experienced team from world-class companies (e.g. Goldman Sachs, TikTok, Accenture, FPT, etc.)
• Work with a cross-disciplinary team blending education, neuroscience, and AI.
• Freedom to explore, test hypotheses, and publish findings that advance learning analytics.
• See your models directly influence real students growth and learning outcomes.
• Collaborative, purpose-driven team culture.
• Competitive salary, benefits, and growth opportunities as the company scales.
• Employees are entitled to 14 days annual leave, 6 days sick leave, 6 days hospitalisation leave, 2 days compassionate leave.