Certificate programme
Data Scientist
Where the analyst describes what happened and why, the data scientist builds models that anticipate what will happen and automate a decision. Predicting school dropout, estimating the risk of payment default, segmenting customers, detecting fraud, automatically classifying thousands of open-ended answers: these are tasks that belong to this profession.
- Duration
- 5 weeks, 10 sessions, 50 hours
- Format
- In person or online
- Credential
- Training certificate
Tuition175,000 XOF
By the end of the programme, you will be able to
- Frame a business problem as a learning problem and choose the right family of models
- Prepare and clean a dataset without leaking information
- Build explanatory variables that carry real signal
- Train and compare regression and classification models
- Evaluate a model with the right metrics and detect overfitting
- Interpret a model and explain a prediction to a non-technical decision-maker
- Put a model in people's hands as a demonstration, and use generative AI as a working tool
- Carry out a complete project and present it to a group
Who it is for
Target audience
The certificate is for people who want to go beyond descriptive analysis: working data analysts, statisticians, economists, engineers, graduates in mathematics, computer science, statistics or economics, researchers who want to industrialise their processing.
Prerequisites
Together with econometrics, this is the most demanding course in the catalogue. Entry requirements are strict:
- Hold at least a three-year degree (BAC+3) in a quantitative field: mathematics, statistics, computer science, economics, engineering, physics
- Master the basics of Python: variables, lists, dictionaries, conditions, loops and functions, as well as handling a data frame with pandas. The Waaldé Impact Data Analyst certificate or Python course meets this requirement
- Master the concepts of inferential statistics: distribution, sampling, hypothesis testing, correlation, linear regression
- Have a laptop with at least 8 GB of RAM, and an internet connection at home
A fifteen-minute technical interview is held before any enrolment. It is not meant to exclude, but to avoid a costly enrolment for someone who would drop out by the third week. Candidates judged insufficiently prepared are directed to the Data Analyst certificate, which is the previous step.
Teaching method
The certificate runs over five weeks, with two sessions a week, on Saturday and Sunday from 8 a.m. to 1 p.m.: four weeks of teaching and one week of project presentations — ten sessions and 50 supervised hours. Personal work between sessions is estimated at four hours a week.
A fifteen-minute technical interview comes before any enrolment. Its purpose is not to exclude but to spare someone an expensive enrolment they would abandon by the third week: a candidate judged insufficiently prepared is pointed to the Data Analyst certificate, which is the step before. Each cohort is limited to fifteen participants.
The course is built on what the field shows: a model's performance depends far more on the quality of the data and the relevance of the variables built from it than on the choice of algorithm. Preparation, feature engineering and honest evaluation therefore take up most of the time. Every session has a "going further" section covered in the course material, for those who want to go beyond the taught programme.
Every participant leaves with:
- The complete course material in PDF, module by module
- All the notebooks used in class, commented line by line
- The practice datasets and their solutions
- A structured project template — folder layout, environment, experiment tracking — ready to clone
- A cheat sheet for scikit-learn, pandas and evaluation metrics
- A pre-deployment checklist
- A guide to preparing for technical interviews
All the tools used are free and open source.
Detailed program
10 sessions · 50 h of teaching
Week 1 — Foundations and data preparation
2 sessions · 10 h- What a data scientist does, and what sets the role apart from the analyst's
- The cycle of a data science project, from the business problem to the decision
- The main families of problems: regression, classification, segmentation
- The working environment: Anaconda, Jupyter, Google Colab
- Python and pandas refresher: reading, selecting and transforming a data frame
- Git and GitHub: publishing your work from day one
- Exercise: frame four real business problems as learning problems and create your first GitHub repository
- Going further, in the course material: the mathematical notions that matter (vectors, mean, derivative), explained through their use
- Exploratory analysis aimed at modelling: distribution of the target variable and class imbalance
- Visualisation with matplotlib and seaborn
- Missing values and outliers: spotting them, understanding them, deciding
- Data leakage, the mistake that produces brilliant and unusable models
- Split before you prepare: the training and test split rule
- Exercise: clean a dataset containing a deliberate leak and write the exploration report
- Going further, in the course material: transformations (logarithm, normalisation) and the statistical tests worth running before modelling
Week 2 — Feature engineering and first models
2 sessions · 10 h- What makes an explanatory variable useful — often more decisive than the choice of algorithm
- Encoding categorical variables: dummy variables and ordinal encoding
- Date-based variables: year, month, length of time
- Aggregate variables: group means and running totals
- scikit-learn pipelines, so that a transformation is never applied twice
- Exercise: build about ten variables on a real dataset and measure what each adds
- Going further, in the course material: ratios and interactions between variables
- The shared logic of scikit-learn: fit, predict, evaluate
- Linear regression to predict a numerical value
- Logistic regression and the decision tree for classification
- Cross-validation: why a single split is not enough
- Exercise: train and compare three models on the same prediction problem
- Going further, in the course material: k-nearest neighbours and why scaling matters
Week 3 — Evaluating and improving models
2 sessions · 10 h- Regression metrics: mean absolute error, RMSE and R²
- Classification metrics: accuracy, precision, recall, F1 and reading the confusion matrix
- Overfitting and underfitting: recognising the bias–variance trade-off on the curves
- Ridge and lasso regularisation: what it corrects
- Exercise: evaluate two models with different metrics and justify the one you keep
- Going further, in the course material: ROC curves and precision–recall curves
- Random forests and gradient boosting: principle and use
- Tuning hyperparameters with grid search and cross-validation
- Interpreting a model through variable importance
- Explaining a prediction to a non-technical decision-maker
- Exercise: improve the week's best model and write the justification for your choice
- Going further, in the course material: unsupervised learning with k-means clustering
Week 4 — Personal project and showcasing it
2 sessions · 10 h- Saving a trained model and loading it again for new data
- Building a small demonstration app with Streamlit
- Generative AI as an assistant: code generation, documentation — and its limits
- Responsible use: verifying, documenting and citing your sources
- Exercise: produce a working demonstration of your model
- Choosing and scoping the subject: problem, data, success criterion
- Organising the GitHub repository: structure, README, experiment tracking
- Writing the project report and its conclusions
- Preparing the ten-minute presentation
- Exercise: individual review of the project with the trainer
Week 5 — Portfolio project presentations
2 sessions · 10 h- A ten-minute presentation of each project, followed by five minutes of questions
- Structured feedback from the trainer and from peers: strengths and areas to improve
- Exercise: every participant presents their portfolio project
- Presentation of the remaining projects, in the same format
- Collective review and advice on showcasing the portfolio to recruiters
- Final assessment and award of certificates
Assessment and certification
The Data Scientist certificate is awarded on an admission average out of 20, made up of three components:
- The certification exam (50%) — 25 questions drawn at random from a bank of more than eighty, in 90 minutes. Two attempts are allowed, seven days apart, and the better one counts.
- The certification project (30%) — defended in the tenth session.
- Attendance (20%) — recorded session by session. An absence announced at least three days in advance counts as half; an unannounced absence does not count.
The admission average is 12 out of 20. A mark below 8 out of 20 in the exam is eliminatory: neither the project nor attendance can make up for it. As soon as the average is reached, the certificate is issued automatically and becomes available in the participant's account.
The certificate is issued by Waaldé Impact, signed by the trainer, and states the course title, the 50 hours, the period, the list of modules completed and the subject of the certification project presented. Each certificate carries a unique number and a verification code.
Participants who do not reach the average receive a certificate of attendance. They may retake the exam and present their certification project with the following cohort, at no extra cost.
Upcoming sessions
Next dates being prepared
Leave your email: we will let you know as soon as registration opens.
Frequently asked questions
175,000 XOF per participant. Training for a team or an organisation is quoted on request.
5 weeks, 10 sessions, 50 hours.
Yes: you can attend onsite or online, as you prefer.
Together with econometrics, this is the most demanding course in the catalogue. Entry requirements are strict: hold at least a three-year degree (BAC+3) in a quantitative field: mathematics, statistics, computer science, economics, engineering, physics ; master the basics of Python: variables, lists, dictionaries, conditions, loops and functions, as well as handling a data frame with pandas. The Waaldé Impact Data Analyst certificate or Python course meets this requirement ; master the concepts of inferential statistics: distribution, sampling, hypothesis testing, correlation, linear regression ; have a laptop with at least 8 GB of RAM, and an internet connection at home ; a fifteen-minute…
Yes: participants who meet the completion requirements receive a Waaldé Impact certificate with a unique number that employers can verify online.
Fill in the online enrolment request: no payment is taken on the site. The firm reviews your application, then contacts you to confirm your place and the payment terms. No session is scheduled yet: leave your email on this page to be told when registration opens.
Join the next cohort
Cohorts are deliberately small to guarantee genuine individual support. Reserve your place or ask us your questions.