MSc Data Science
A cutting-edge one-year postgraduate programme that trains the next generation of senior data scientists and AI professionals. With a 98% placement rate and a defining industry capstone, graduates enter the workforce with both deep technical expertise and real project experience.
Programme Overview
The MSc Data Science delivers advanced training across the full data science stack — from distributed cloud computing and database engineering to deep learning and NLP. Semester 1 covers core technical modules; Semester 2 adds specialisations in NLP, data privacy, and optimisation for ML. The programme concludes with a 12-week capstone project embedded within a partner organisation — a live professional experience that frequently results in direct graduate hires.
Partner organisations for capstone projects include firms across fintech, healthcare analytics, e-commerce, climate science, and government data agencies.
What You'll Learn
- Deep Learning: CNNs, RNNs, transformers, and large language model fine-tuning
- Advanced Databases & Cloud: Distributed systems, AWS/GCP, data lake architecture
- Predictive Analytics: Forecasting, time series, regression with model deployment
- NLP & Text Mining: Tokenisation, embeddings, sentiment, named entity recognition
- Data Engineering: Pipeline design, orchestration (Airflow), streaming (Kafka)
- Optimisation for ML: Gradient descent variants, hyperparameter search, AutoML
- Data Privacy & Governance: Differential privacy, federated learning, data stewardship
- Capstone Project: Industry-embedded 12-week applied project with real deliverables
Deep Learning
Architecture design, training techniques, regularisation, and deployment of deep neural networks for vision and language tasks.
Advanced Databases & Cloud
Distributed SQL and NoSQL systems, cloud-native data architecture (AWS Redshift, BigQuery), and data lake design.
Applied Statistics
GLMs, mixed models, and advanced regression techniques applied to real datasets from healthcare, finance, and social science.
Predictive Analytics
Forecasting methods — exponential smoothing, ARIMA, Prophet, gradient-boosted trees — with production deployment.
Distributed Computing
Apache Spark, Hadoop, Dask, and Kubernetes for scalable data processing across cloud clusters.
NLP & Text Mining
Classical NLP pipelines through to transformer models (BERT, GPT) — fine-tuning and deploying LLMs for domain tasks.
Data Engineering
Building robust ETL/ELT pipelines using Apache Airflow, dbt, and modern data stack tools for analytics engineering.
Industry Capstone
12-week embedded project with a partner organisation, delivering an analytics product with full documentation and presentation.
Click any course to view its objective and learning outcomes.
DSC 501 Mathematical Foundations for Data Science +
Objective
To consolidate the linear algebra, calculus and probability essential for advanced ML.
Learning Outcomes
- Apply matrix calculus to backpropagation.
- Use SVD and PCA for dimensionality reduction.
- Apply convex analysis to optimisation.
- Use measure-theoretic probability where appropriate.
- Implement vectorised computations efficiently.
DSC 502 Statistical Learning Theory +
Objective
To analyse the theoretical foundations of supervised learning.
Learning Outcomes
- Apply VC dimension and Rademacher complexity.
- Use PAC learning bounds.
- Apply concentration inequalities.
- Analyse generalisation gap.
- Discuss bias-variance trade-off rigorously.
DSC 503 Advanced Machine Learning +
Objective
To master state-of-the-art supervised and unsupervised learning algorithms.
Learning Outcomes
- Apply gradient boosting and ensemble methods.
- Use kernel methods and SVMs.
- Apply unsupervised clustering and dimensionality reduction.
- Use semi-supervised and active learning.
- Tune hyperparameters with Bayesian optimisation.
DSC 504 Deep Learning +
Objective
To design, train and deploy modern neural networks at scale.
Learning Outcomes
- Implement CNNs, RNNs and Transformers.
- Apply transfer learning effectively.
- Use distributed training on GPUs.
- Apply self-supervised pretraining.
- Deploy models for inference.
DSC 505 Big Data Engineering +
Objective
To process data at scale using modern distributed systems.
Learning Outcomes
- Apply Spark for distributed processing.
- Use Kafka and Flink for streaming.
- Build data lakehouse architectures.
- Apply orchestration via Airflow.
- Manage cloud data infrastructure.
DSC 506 NLP & Text Analytics +
Objective
To analyse and model text data with modern transformer-based methods.
Learning Outcomes
- Apply tokenisation and embeddings.
- Train transformer models for classification and generation.
- Use LLMs via fine-tuning and prompting.
- Apply RAG architectures.
- Evaluate NLP systems with appropriate metrics.
DSC 507 Computer Vision +
Objective
To build computer vision models for classification, detection and segmentation.
Learning Outcomes
- Apply CNNs to image classification.
- Use object detection (YOLO, Faster R-CNN).
- Apply semantic and instance segmentation.
- Use vision transformers.
- Apply self-supervised vision methods.
DSC 508 Reinforcement Learning +
Objective
To formalise sequential decision-making and apply RL algorithms.
Learning Outcomes
- Apply Markov Decision Processes.
- Use Q-learning and policy gradients.
- Apply actor-critic methods.
- Use deep RL (DQN, PPO).
- Apply RL to real-world problems.
DSC 509 MLOps & Production +
Objective
To deploy and maintain ML systems in production.
Learning Outcomes
- Build CI/CD pipelines for ML.
- Use feature stores and model registries.
- Apply monitoring and observability.
- Use containerisation and Kubernetes.
- Manage model versioning and rollback.
DSC 510 Data Ethics & Fairness +
Objective
To address ethical, legal and societal implications of data science.
Learning Outcomes
- Audit ML systems for bias.
- Apply differential privacy.
- Use federated learning.
- Discuss interpretability methods.
- Comply with GDPR and similar regulations.
AND OR NOT XOR -> <->
DSC 511 Advanced Causal Inference +
Objective
To estimate causal effects using modern statistical methods.
Learning Outcomes
- Apply potential-outcomes framework.
- Use propensity scores and matching.
- Apply instrumental variables.
- Use difference-in-differences.
- Apply causal ML methods.
DSC 512 Master's Capstone +
Objective
To deliver an industry-grade end-to-end data science project.
Learning Outcomes
- Frame a complex business problem.
- Build production-quality ML systems.
- Communicate with stakeholders.
- Document for handoff and reproducibility.
- Present to a technical and lay audience.
Senior Data Scientist
Lead data science squads at scale-ups and enterprise firms, owning end-to-end ML model lifecycle and business impact.
ML Engineer
Design and productionise ML systems using MLOps frameworks, serving models at scale with monitoring and retraining pipelines.
AI Product Manager
Bridge technical AI capabilities and business strategy, defining AI product roadmaps at technology companies.
Chief Data Officer
Executive-level leadership of data strategy, governance, and analytics capability within large organisations.
Data Architect
Design enterprise-scale data platforms, warehouses, and lake architectures that support thousands of analytical users.
Research Scientist
Conduct applied ML research at corporate labs (Google Brain, Meta AI) or academic research centres.
Why D'Math University
98% Placement Rate
Our dedicated careers team, alumni network, and capstone partnerships deliver extraordinary graduate employment outcomes.
Full-Stack Data Training
From statistical modelling to cloud infrastructure and deep learning — graduates are equipped across the entire data science spectrum.
Live Industry Capstone
Real problem, real data, real client — the capstone is the centrepiece of the programme and a launching pad for careers.
Expert Mentorship
Each student is paired with an industry mentor from our partner network for guidance throughout the programme.
September and January intakes — early applications receive priority capstone placement