Data preparation
Cleaning, joining and shaping data before any modelling begins.
Training Programme
Data Science, AI & Machine Learning
From data preparation to deployed models, with the evaluation discipline that separates a demo from a production system.
Overview
What this gives you
The hard part of machine learning is not fitting a model. It is establishing whether the model is predicting something real, whether it will still work when the data moves, and whether anyone will act on its output. Those are the skills this programme is built around.
Participants work end to end: clean a real dataset, build and evaluate a model against a held-out set, put it behind an interface, and then monitor it. The governance and evaluation material is included because models that reach production without it create risk rather than value.
- Supplied, installed and supported by Quridge Global Solutions from Bengaluru
- Written quotation with GST-inclusive pricing, warranty and AMC options
- Documentation, operator training and a named point of contact for support
At a glance
What you get
Evaluation first
Held-out evaluation and baselines agreed before training.
Model families
Supervised, unsupervised, deep learning, NLP and computer vision.
Deploy and monitor
MLOps, model deployment, drift detection and governance.
Specification
Technical details
| Module | Content |
|---|---|
| Python for data analysis | Core Python, NumPy, pandas and exploratory analysis |
| Statistics and probability | Descriptive statistics, distributions, inference and testing |
| Machine learning | Supervised and unsupervised learning, feature engineering and evaluation |
| Deep learning and computer vision | Neural networks, CNNs, transfer learning and vision tasks |
| Generative AI and LLM applications | Prompt design, retrieval, evaluation and guardrails |
| MLOps | Deployment, versioning, monitoring, drift detection and retraining |
Where it is used
Typical applications
The situations this is most often specified for, based on what we are asked to quote most often.
- Analysts moving into data science
- Engineers building their first model
- Teams evaluating an AI vendor
- Leadership needing to assess AI claims
FAQ
Frequently asked questions
Comfort with high-school algebra and some statistics is enough for the core programme. We include a short statistics refresher, and nothing beyond it is assumed.
Yes. Each participant takes a dataset through cleaning, modelling, evaluation and deployment, and leaves with a working model and the code that produced it.
Yes, as a module covering prompt design, retrieval, evaluation and the guardrails that keep a language model application safe enough to put in front of a user.
Related
Other options in this range
Need a recommendation?
Tell us your volume, media, site count and compliance requirement. We will come back with a model, a costed quotation and AMC options within one business day — with no obligation.