Certificate Program in Applied AI — Genvarsity
Machine Learning
Data Science
Generative AI
Neural Networks
LLMs & RAG
SQL & Analytics
Industry Internship
Applied AI
Python & Data Toolkit
Prompt Engineering
Certificate Program Mastery Level: Intermediate 12 Months Total

Certificate Program in Applied AI

Ten months of live Applied AI curriculum covering Machine Learning, Data Science and Generative AI, delivered completely remote, followed by a 2-month guaranteed industry internship with a host company. Designed for second and third year students in four-year programmes — BCA, BTech and equivalent.

10
Month Courseware
255+
Structured Hours
2
Month Internship
Live
Remote Delivery
// programme overview

Ten months of curriculum.
Two months of internship.

The programme runs for twelve months in total. The first ten months are structured Applied AI courseware delivered live and remotely. The final two months are an industry internship at a host company, with a company-issued internship certificate on completion.

📚
10 Months
Structured Courseware
255+ hours of live online sessions covering ML, Data Science and Generative AI — plus activity hours, assignments and projects throughout. Designed to fit alongside your college semester.
Intermediate Mastery
🏢
2 Months
Guaranteed Industry Internship
On completion of the courseware, students are placed with a Genvarsity host company for two months of project work. An internship certificate is issued on the host company's letterhead, independent of Genvarsity's programme certificate.
🎓
7 Modules
End-to-End AI Stack
Seven sequential modules covering Python, mathematics for ML, classical machine learning, deep learning, data science, generative AI, and a capstone project. Each module is led by a practitioner currently working in the field.
// curriculum · 7 modules

Seven modules.
Intermediate level throughout.

Modules are sequenced so each builds on the previous. Python and mathematics precede ML; ML and deep learning precede GenAI. The capstone in month ten integrates two or more modules into a single deployed project.

255+ hours Structured live sessions
+ activity & project hours
01
Python & Data Toolkit 30 hours · Weeks 1–3
Foundation

Proficiency in Python and the core data manipulation libraries is a prerequisite for every subsequent module. This module covers both at working level. Students who already have Python experience move through this module faster; sessions are structured to accommodate varied starting points.

Python & Programming
Python data types, control flow, functions and scope
List comprehensions, lambda functions, iterators
Object-oriented programming — classes, inheritance, encapsulation
File handling — reading and writing CSV, JSON, text
Working with APIs — requests, parsing JSON responses
Exception handling and debugging strategies
Virtual environments and package management (pip)
Data Toolkit
NumPy — arrays, vectorised operations, broadcasting
Pandas — DataFrames, Series, indexing and slicing
Data cleaning — missing values, duplicates, type conversion
GroupBy, merge, pivot and reshape operations
Matplotlib and Seaborn — foundational visualisations
Jupyter and Google Colab — professional workflow
Git basics — version control for data projects
Toolkit
Python 3.x NumPy Pandas Matplotlib Seaborn Jupyter Git
02
Mathematics for AI 20 hours · Weeks 4–5
Foundations

Covers the linear algebra, probability, statistics and calculus concepts that underpin ML algorithms. Treatment is applied rather than proof-based — the goal is working understanding of why models behave as they do, not mathematical derivation. Students completing this module can interpret loss functions, read gradient descent diagrams and reason about model optimisation.

Linear Algebra
Scalars, vectors and matrices — what they represent
Matrix multiplication and dot products
Transpose, inverse and identity matrices
Eigenvalues and eigenvectors — intuitive understanding
How linear algebra underlies data transformations in ML
Probability, Stats & Calculus Intuition
Probability fundamentals — events, distributions, Bayes theorem
Descriptive statistics — mean, variance, standard deviation
Normal distribution, central limit theorem
Correlation versus causation
Gradient intuition — what a derivative is, what it tells you
Loss functions — MSE and cross-entropy explained
Optimisation intuition — why we descend gradients
03
Machine Learning — Core 55 hours · Weeks 6–11
Core ML

Classical supervised and unsupervised ML at working depth — covering algorithm selection, implementation, evaluation and diagnosis. Students work through each algorithm with explicit attention to failure modes, not only successful training runs. scikit-learn Pipelines are introduced as the standard for reproducible, production-oriented ML code.

Supervised Learning
Linear regression — OLS, regularisation (Ridge, Lasso, ElasticNet)
Logistic regression — binary and multiclass
Decision trees — splitting criteria, depth, pruning
Random Forests — bagging, feature importance
Gradient Boosting — XGBoost, LightGBM in practice
Support Vector Machines — linear and kernel
Naive Bayes and k-Nearest Neighbours
Evaluation, Features & Unsupervised
Train/test split, cross-validation, stratification
Evaluation metrics — accuracy, precision, recall, F1, ROC-AUC
Confusion matrix interpretation and business alignment
Bias-variance tradeoff and overfitting diagnosis
Feature engineering — encoding, scaling, imputation
Feature selection — filter, wrapper, embedded methods
Clustering — K-means, DBSCAN, hierarchical
Dimensionality reduction — PCA applied
Hyperparameter tuning — GridSearchCV, basics of Optuna
scikit-learn Pipelines for reproducible ML
Toolkit
scikit-learn XGBoost LightGBM SHAP Optuna Plotly
04
Deep Learning — Essentials 40 hours · Weeks 12–16
Deep Learning

Covers neural network architecture, training dynamics, and the transformer model — the architecture underlying all current large language models. Transfer learning is covered at applied level: students fine-tune pre-trained models from Hugging Face on custom datasets. Production-level training infrastructure (distributed training, MLOps tooling) is out of scope at this level and covered in the IIP.

Neural Network Fundamentals
Perceptrons and multi-layer networks — forward pass
Activation functions — ReLU, sigmoid, softmax
Backpropagation — intuitive walkthrough
Optimisers — SGD, Adam, learning rate schedules
Regularisation — dropout, batch normalisation, weight decay
Training dynamics — overfitting, early stopping
PyTorch basics — tensors, autograd, training loop
Architectures & Transfer Learning
CNNs — convolution, pooling, classic architectures (ResNet concept)
Sequence models — RNNs and LSTMs (awareness level)
Attention mechanism — what it solves and why it works
Transformer architecture — encoder, decoder, self-attention
Why transformers replaced RNNs — intuitive explanation
Transfer learning — using Hugging Face pre-trained models
Fine-tuning a pre-trained model on a custom dataset
Toolkit
PyTorch Hugging Face TensorFlow (awareness) Google Colab
05
Data Science — Applied 40 hours · Weeks 17–21
Data Science

Covers statistical inference, experimental design, SQL at working level, and business intelligence tooling. Students completing this module can frame a business question as an analytical problem, query a relational database, run and interpret an A/B test, and produce a Power BI dashboard that a non-technical stakeholder can use.

Analysis & Statistics
Exploratory data analysis — univariate and bivariate
Distribution analysis and outlier identification
Statistical inference — hypothesis testing, p-values
Confidence intervals and effect sizes
A/B testing — design, execution and interpretation
Correlation analysis and multicollinearity
Time series basics — trend, seasonality, ARIMA, Prophet
SQL, Visualisation & Communication
SQL — SELECT, JOINs, GROUP BY, window functions, CTEs
Query optimisation fundamentals
Working with databases — PostgreSQL, BigQuery basics
Chart selection and visualisation principles
Power BI — data model, DAX basics, dashboard design
Plotly and Streamlit for analytical apps
Communicating findings — structuring analytical narratives
Toolkit
SQL PostgreSQL Power BI Plotly Streamlit Prophet
06
Generative AI & LLMs 50 hours · Weeks 22–28
Generative AI

Covers LLM architecture, prompt engineering, and GenAI application development. Students work through commercial LLM APIs, build a functional RAG pipeline, and implement a basic AI agent using the ReAct pattern. Fine-tuning is covered at awareness level — the decision framework for when to fine-tune versus prompt-engineer versus RAG — with hands-on fine-tuning reserved for the IIP. Students complete this module able to build and deploy a functional LLM-powered application using current tooling.

LLM Foundations & Prompt Engineering
How LLMs work — pretraining, SFT, RLHF (conceptual)
Tokenisation, context windows, embedding spaces
Working with LLM APIs — OpenAI, Anthropic Claude, Gemini
Prompt engineering — system prompts, few-shot, chain-of-thought
Structured outputs — JSON mode, function calling
Output validation with Pydantic
Inference parameters — temperature, top-p, max tokens
Prompt failure modes — hallucination, refusal, drift
RAG, Agents & Applications
RAG pipeline — chunking, embeddings, vector stores, retrieval
Embedding models — what they are, how to select
Vector databases — Chroma, Pinecone in practice
Hybrid retrieval — semantic plus keyword search
LangChain basics — chains, prompts, output parsers
AI agents — ReAct pattern, tool use, basic agent loop
Building an LLM-powered application end to end
Fine-tuning awareness — what it is and when to use it
Toolkit
OpenAI API Anthropic Claude LangChain Chroma Pinecone Pydantic Streamlit
07 · Capstone Project — 30 hours · Weeks 29–32
End-to-End Applied AI Project
Students select a real-world problem, build a complete solution combining at least two modules — typically ML plus GenAI — and deploy it with a working interface. The capstone is the portfolio piece they carry into the internship and beyond. Reviewed by an industry panel before internship placement.
Problem framing & dataset selection Full ML + GenAI pipeline Streamlit or Gradio deployment Technical documentation Industry panel review
// the internship

Two months.
Host company placement. Company-issued certificate.

On completing the ten-month courseware and the capstone project, students are placed with a Genvarsity host company for two months of project work. The internship certificate is issued by the host company, not by Genvarsity, and constitutes a verifiable record of employment-equivalent engagement.

Guaranteed placement on completion
Placement process begins before courseware ends.
Internship matching begins in month nine, while the courseware is still running. By the time the capstone is completed and reviewed, the host company, supervisor and project brief are confirmed.
Mo 1–10Courseware + capstone project. Internship matching begins month nine.
End Mo 10Capstone reviewed. Internship placement confirmed. Company and brief assigned.
Mo 11–12Two months at the host company. Real project work. Company-issued internship certificate.
CompletionGenvarsity certificate + host company internship letter. Ready for the job market or the IIP.
On stipends: Internship stipend, if any, is entirely at the sole discretion of the host company and does not form part of this programme's commitment. Stipend arrangements, where applicable, are communicated at the time of placement.
01
Matched to your project area
Placement is matched to the domain covered in the capstone — ML, Data Science or Generative AI. The host company receives a brief on the student's project work prior to the internship commencing.
02
Certificate on company letterhead
The internship certificate is a company document — not a Genvarsity certificate. It records your engagement and contribution in the company's name.
03
Verifiable work experience
Two months of company engagement documented on the host's letterhead, with supervisor details. The certificate can be verified directly with the issuing company.
04
Gateway to the IIP
Certificate graduates who proceed to the Industry Immersion Programme in their final year receive a progression discount and enter at advanced depth from month one.
// programme credentials

Programme certificate.
Company internship letter.

Students who complete the full twelve months — ten months courseware and two months internship — receive two separate documents, issued independently by Genvarsity and the host company respectively.

01
Genvarsity Programme Certificate
Certificate in Applied AI from Genvarsity, confirming completion of the 10-month curriculum covering Machine Learning, Data Science and Generative AI at intermediate mastery level, including capstone project review.
02
Host Company Internship Letter
Issued on the host organisation's letterhead — a real company document recording two months of project engagement, the student's contribution, and the supervisor's details. This is a verifiable employment record.
Position at graduation
Applied AI skills at intermediate level, a deployed capstone project, and two months of verifiable company experience. Students who proceed to the IIP in their final year add a twelve-month company engagement on top of this, prior to graduation.
// admissions · 2026

Applications for the
2026 cohort are open.

Speak with a Genvarsity counsellor to discuss programme fit, academic scheduling and the admission process.

Talk to a Counsellor
Collaborative Training Platform Genvarsity operates as a collaborative training platform. Classes and live sessions are delivered by working professionals from our partner companies — bringing real-world practice into every programme.
AI Tutor — Powered by Genbyte AI The Genvarsity AI Tutor is proprietary technology developed and owned by Genbyte AI. All intellectual property rights in the AI Tutor are reserved by Genbyte AI.
Internship stipend, if any, is at the sole discretion of the host organisation and does not form part of the programme commitment. Genvarsity does not guarantee or standardise stipend amounts. Fee waiver available for students who have lost both parents or whose parents are unable to support their education — handled through the counselling process. © 2026 Ensynapse Technique Pvt. Ltd. All rights reserved.