Programme

Generative AI for Business

Duration	3 Months
Contact hours	240 hours
Level	Beginner
Modules	6
Topics	24 (16 practical)
Fee	PKR 60,000

What the programme includes

Live classes with a named trainer, on a published schedule
Recordings of every session, so a missed class is not a lost one
Marked assignments with written feedback
Project work reviewed and verified by your trainer
A final assessment
A completion letter stating your attendance, results and verified work — independently verifiable by any employer

Syllabus

1. What it can and cannot do for a business

30 hours

Separating the useful from the sold.

Learned from examples, not programmed
The one idea that explains every strength and failure that follows.
The confident wrong answer
No change in tone when it does not know. What that means for anything customer-facing.
Where it earns money today
Drafting, summarising, classifying, answering the same questions. Repetitive, high volume, human-checked.
Judging a vendor's claim hands on
Learned from what, to do what exactly. If they cannot answer, there is usually nothing there.

2. Getting reliable output

40 hours

The practical skill, without the mythology.

Answering the questions a colleague would have asked hands on
The fix for nine out of ten poor results, and it takes twenty seconds.
Showing an example beats describing one hands on
Two samples of the output you want, and why it works.
Fixing the output format hands on
Getting a table, a list or a fixed structure so the result can be used rather than retyped.
Building a reusable brief hands on
Turning a good prompt into something a whole team runs.

3. Writing, selling and serving

50 hours

Applying it to the work a business already does.

Marketing and sales copy, in Urdu and English hands on
Keeping a brand voice instead of sounding like everyone else.
Proposals, quotations and contracts
Drafting fast, and the clauses a human must always read.
Customer support answers hands on
Drafting replies from your own policies rather than from the model's imagination.
Meeting notes and reports hands on
Turning a recording into actions with owners and dates.

4. Documents and data

40 hours

The highest-value use in most Pakistani firms.

Reading invoices, forms and CVs into structured fields hands on
Where hours genuinely disappear from an office.
Summarising long documents faithfully hands on
Checking a summary against the source before acting on it.
An assistant over your own documents hands on
Answering from your policy and price list, and refusing what is not there. Set up without code.
Spreadsheets and analysis hands on
Cleaning, categorising and explaining data.

5. Images, audio and video

40 hours

Creative production for a small marketing budget.

Image generation and editing for campaigns hands on
Product mockups and social creative, and the limits worth knowing.
Voice and video hands on
Narration in Urdu and English, subtitles and short-form editing.
Brand consistency
Keeping a recognisable look when the tool defaults to generic.
Disclosure and likeness
What must be declared, and why an undisclosed synthetic endorsement is not recoverable.

6. Rolling it out without an incident

40 hours

The governance a business needs before staff start pasting.

What must never be pasted into a public tool hands on
Customer records, CNIC numbers, contracts under NDA, credentials. The rule, written for staff.
An AI use policy on one page hands on
Drafted in class, for your own organisation.
Who checks the output, and when
Deciding where a human signature is required before anything leaves the building.
Cost, licensing and vendor lock-in
What subscriptions actually cost at team scale, and what happens if you stop paying.

Against the national standard

Pakistan's National Vocational & Technical Training Commission publishes a course-contents document for the trades it funds. This programme is mapped to AI, Machine Learning & Deep Learning — 12 weeks, 4 hours a day, 240 contact hours.

Published by NAVTTC and reproduced here so you can check our syllabus against it. We teach it to those hours: 240 contact hours over 12 weeks, the same as the standard. Read the NAVTTC document.

11 modules in the national standard

Introduction Linux Shell Scripting Fundamentals Python Fundamentals — week 1, 20 hours
Introduction to AI Motivational Lecture · Course Introduction Job market Course Applications Work ethics Survey of career opportunities Survey of industry requirements for each career path · Software Installation (Anaconda, VSCode, PyCharm, etc.) · Introduction to Debian Basic Commands: pwd, cd, ls, cat, sudo, man, redirection, mkdir, rm, rmdir, cp, mv · file, reading, cat, more, less, head, alias...
Overview — week 2, 20 hours
Python Fundamentals Implementation of OOP Principals in Python Descriptive Statistics and Probability · Functions Functions and variable scope Lambda expression Map and Filter Inner/Nested functions · File Handling Exception Handling · Classes and Objects Instance Variables and Methods Class Variables and Functions Constructors and Destructors · Inheritance Multilevel Inheritance Hierarchical Inhe...
Descriptive Statistics and Probability Overview Python Support Libraries for Exploratory Data Analysis - NUMPY — week 3, 20 hours
Correlation Coefficient · Univariate, bivariate and multivariate plots · Probability · Joint, Marginal and Conditional probability · Probability Distributions · Discrete and Continuous probability distributions Bayesian Probability · Introduction to Numpy · Creating Numpy Arrays (from Python list, from built-in methods, from random) Array Attributes and Methods (reshape, max, min, argmax, argmin,....
Python Support Libraries for Exploratory Data Analysis - Pandas - Seaborn — week 4, 20 hours
Merging, Joining, and Concatenation (inner, outer, right and left joins) · GroupBy Discretization and Binning Operations on DataFrames Data output/saving Pandas for Plotting (area, bar, density, hist, line, scatter, barh,... · Introduction to Seaborn · Distribution Plots distplot jointplot (pairplot, rugplot, kdeplot) · Categorical Data Plots factorplot, boxplot, violinplot, stripplot, swarmplot,...
Machine Learning-I — week 5, 20 hours
Multivariate Linear Regression · Polynomial Regression · Logistic Regression (Binary Classification) · Logistic Regression (Multiclass Classification) · Code practice
Natural Language Processing — week 6, 20 hours
Introduction to Natural Language Processing · Syntax, Semantics, Pragmatics, and Discourse NLP curves and future directions · Data pre-processing for NLP Introduction to NLTK/SpaCy Noise removal (stopwords, punctuation, etc) · Word and sentence tokenization Word segmentation Stemming Text normalization Regular expression for string parsing · POS tagging NER tagging Chunking and Chinking Lemmatizat...
Deep Learning I — week 7, 20 hours
Boosting · MLP Feed Forward Neural Network Forward and backward passes Nonlinearity: Activation functions Cross-Entropy Computational graph and... · Introduction and implementation of neural networks using appropriate deep learning API of choice (TensorFlow, PyTorch, Keras) · Convolutional Neural Network (CNN) 2D CNN for image classification · 1D CNN for text document classification · Code Practic...
Deep Learning II — weeks 8, 9, 40 hours
Recurrent Neural Networks (RNNs) · Long-Short-Term-Memory Networks (LSTM) · LSTM Code Practice · Gated Recurrent Unit Networks · GRU Code Practice · Word Embeddings Word2vec Continuous BOW Continuous Skip-gram · Gensim and Custom Embedding Training · Sequence Models
Bi-Directional LSTM/RNN in Sequence Models — week 10, 20 hours
Deep Learning II Employable Project / Assignment (2 weeks, 11-12) in addition of regular classes. OR On job training (2 weeks) · Attention Mechanism in Models · Selection of Project, architecture discussion, preparation. Guidelines to the Trainees for selection of employable project like final year project... · https://technofizi.net/best-computer- science- and-engineering-cse-project- topics-idea...
MS Azure AI Service — week 11, 20 hours
Selection of Microsoft Azure AI Service Selection the appropriate service for a vision solution Selection the appropriate service for a language... · Selection the appropriate service for a decision support solution · Selection the appropriate service in Cognitive Services for a speech solution Selection the appropriate Applied AI services · Configuring Security for Microsoft Azure AI Service Mana...
Process videos — week 12, 20 hours
Microsoft Azure Natural Language Processing (NLP) Solutions Implementation Analyze text Process speech Translate language · Build and manage a language understanding model Create a question answering solution · Build and manage a language understanding model · Microsoft Azure Knowledge Mining Solutions Implementation · Microsoft Azure Conversational AI Solutions Implementation · Task Title · Descr...

Before, and after

Before you start

No technical background required
This course involves no programming at all.
Free: ChatGPT for Beginners
A useful warm-up, not required.

Where it leads

AI Automation & n8n
For making these processes run without a person starting them.
Digital Marketing
If most of your interest turned out to be marketing.

Roles this prepares you for

Using AI in your existing role
The honest and most common outcome, and a good one — this course is designed for it.
AI Solutions Associate
Helping a team adopt these tools.
Operations / Admin Lead
Owning the processes this course improves.

These are the roles the programme is aimed at. BvLogic Academy does not guarantee employment, an income, or a timeframe — and you should be careful with any institute that does.

Other courses

Full Stack Development · 3 Months
AI & Generative AI · 3 Months
Digital Marketing + AI · 3 Months
Python Development · 3 Months

See all courses →

Interested in this programme?

Try the subject free first — there is no reason to pay before you know you like it.

Browse the free courses Take the career test