Generative AI, ChatGPT 5, Copilot & AI Agents Mastery 2026

Meet instructor Ryan Ahmed and explore the course outline for generative artificial intelligence, ChatGPT, copilot and AI agents for business, covering fundamentals, practical demos, and three-part structure.

Learn by doing as you build and apply state-of-the-art AI models, solve practice opportunities, engage in Q&A, complete all modules on Copilot and ChatGPT topics, and earn your certificate.

Explore the fundamentals of generative artificial intelligence and the key tools like ChatGPT and Copilot, plus the history, AGI, artificial superintelligence, and the future of AI.

Discover the fundamentals of artificial intelligence and generative AI, including how AI enables recommendations and autonomous driving. See how prompts generate novel content such as text, images, code, and more.

Explore the fundamentals of ChatGPT and Microsoft Copilot, powered by GPT-4 for questions, code, and data analysis, plus Copilot’s data-grounded productivity in Word, PowerPoint, and Outlook.

This lecture traces AI history from Eliza to transformers and ChatGPT, then explains future with AI agents and five levels toward AGI and ASI.

Compare popular AI models like GPT-4, Google Gemini, Llama by Meta, and Claude, with generative AI tools, image generation options such as DALL-E, Midjourney, and Stable Diffusion, and AI avatars.

Explore practical capabilities of ChatGPT, Gemini Cloud, and Deep Sea through hands-on prompts, from brainstorming and code generation to financial data analysis, image reasoning, and CAPM problem solving.

Thank you for joining this masterclass on generative AI, copilot, ChatGPT, and agents. Please take five minutes to leave a review; your feedback helps improve future courses.

Explore practical prompts across ChatGPT, Cloud by Anthropic, Google Gemini, and Deep Seek to generate ten bank-focused AI use cases such as AI-powered virtual assistants and automated financial report generation.

This practical demo introduces ChatGPT tools—Dall-E image generation, a search for up to date information, O1 advanced reasoning for multi-step problems, and canvas for real-time collaborative writing and coding.

Explore hands-on demos of ChatGPT tools for image generation with Dall-E, web search, and enhanced reasoning, then use canvas to brainstorm, write, and edit code and case studies.

Explore Microsoft Copilot across PowerPoint, Word, Outlook, and Excel, including prompts, PDF analysis, image generation with DALL-E, and data-driven tasks that automate reports and emails.

Test generative AI capabilities on unstructured product reviews by extracting sentiment and topics, using a sample csv of Eco bottle and Hydra Therm, and output sentiment features with recommendations.

Master how to use generative AI to extract sentiment and topics from real-world customer reviews, build prompts, analyze data in sheets, and generate actionable management recommendations.

Explore ChatGPT five for research, real-time data, and creative writing, using Python code to generate charts, summarize revenues, brainstorm AI agents features, and craft emails and social posts.

Explore how ChatGPT 5 and Sora generate images and videos from prompts, including epic panoramas like the seven wonders, and convert images into editable videos.

Showcases how ChatGPT 5 builds end-to-end applications, converts code between languages, and generates interactive projects like a French learning app with flashcards, quizzes, and pronunciation features.

Learn to use ChatGPT five to extract data from reports, analyze financial statements for liquidity risks, solve problems from images, and generate Python visualizations for datasets.

Master prompt engineering fundamentals by learning how to design context, instructions, input data, and output indicators to tailor responses from ChatGPT 5 and Copilot.

Explore ai agents powered by large language models that plan, remember, and use tools to complete tasks—from booking a flight to deep research and interactive web pages.

See how the AI analyst agent in Microsoft Copilot analyzes sales data, creates ten visualizations, and builds regression models to predict prices with train-test splits and a performance leaderboard.

Learn to create a custom GPT with your data in ChatGPT 5, including uploading knowledge and enabling tools, then build a competitive analysis GPT to compare Lowe's and Home Depot.

Master prompt engineering fundamentals, from zero-shot to few-shot and chain-of-thought, and explore ChatGPT concepts, self-attention, embeddings, OpenAI API basics, and financial data extraction.

Explore artificial intelligence training process with a connected neural network, input, hidden, and output layers, and learn how data and labels drive weight adjustment through supervised training to reduce error.

Build a simple feedforward neural network in Google TensorFlow Playground, train on circle data with two classes, and observe weights adjust over epochs to separate orange and blue.

Explore building an AI with Google TensorFlow Playground by setting an 80/20 training split, adding noise, choosing spiral data, adjusting architecture, and tuning hyperparameters, then evaluate results.

Explore building and tuning a neural network using a spiral dataset in TensorFlow playground, adjusting training/test split, noise, batch size, network architecture, and hyperparameters to assess performance.

Master the fundamentals of prompt engineering, crafting prompts with context, instructions, input data, and output indicators. Discover zero-shot, few-shot, and chain-of-thought prompting to improve AI responses.

Explore how transformers process text with input tokens, self-attention, and multi-layer architectures, using positional encoding to understand context and generate cohesive output.

Explore tensor space dot js visualizations of AI models, where a ResNet brain reveals feature extraction layers and a transformer architecture handles image classification tasks.

Learn how embeddings encode words and images as dense vectors that capture semantic meaning in a space with age, gender, and royalty. See examples with man, woman, king, queen.

Explore the OpenAI API, including access to ChatGPT and GPT-4 for text and code generation, translation, and sentiment analysis. Learn how system and user prompts and temperature shape model behavior.

Leverage generative artificial intelligence to extract balance sheets, income statements, and cash flow statements from 10-Ks and news, visualize results, and export to csv.

Learn to read a 10-K, extract risk factors and financial data, and visualize results with word clouds, charts, and auto-generated PowerPoint slides for Apple.

Extract Walmart's consolidated net income from the 2024 10-K, compare and visualize results, analyze risk factors, summarize the 10-K in a one-page PowerPoint, and analyze past week stock news sentiment.

Practice Opportunity Solution guides you to extract Walmart's consolidated net income for 2022–2024 from the 10-K, assess risk factors, and export visuals, summaries, and sentiment analysis to PowerPoint.

Explore data wrangling and feature engineering with pandas, integrated with ChatGPT, to automate financial planning and analysis tasks across multiple data sources.

Discover how generative ai powers fpna from data wrangling and analysis to forecasting and scenario planning. Visualize data, build predictive models, detect anomalies, and optimize costs to guide strategic decisions.

Explore data wrangling and feature engineering on real-world datasets with generative AI, including cleaning, merging sources, handling missing values, encoding, and preparing data for visualization and modeling.

Learn how the pandas data analysis library powers data manipulation and how ChatGPT uses pandas to generate Python code for cleaning and transforming data frames from CSV, Excel, or SQL.

Read, clean, and handle missing values in two Excel sources in pandas, merge on client ID, then sort, filter, and visualize results in a Jupyter notebook.

Read Excel files into pandas data frames with pd.read_excel, build df_demographics and df_financials, and explore data with head, tail, shape, and info in a Jupyter notebook, handling missing values.

Explore handling missing values in Python with isnull, identify missing data, and fill numerical columns with the mean and categorical ones with the mode, across demographics and financials dataframes.

Learn to merge two pandas data frames on the client ID using pd.merge, creating the merged df clients combined from demographics and financials.

Learn to analyze merged demographics and financial data with pandas: use describe to summarize, filter high net worth clients by portfolio size, and sort results ascending for clear insights.

Visualize data with Python using matplotlib, seaborn, and Plotly; create scatter plots, histograms, bar charts, and a correlation heat map to explore age and investment portfolio size.

Learn feature engineering techniques like imputation, outlier handling, binning, log transforms, and one hot encoding, then apply normalization and standardization to scale features for machine learning models.

Perform data wrangling and feature engineering on a real-world human resources dataset, clean missing values, and apply one-hot encoding to prepare a model that predicts attrition.

Perform data analysis and feature engineering on a human resources dataset by cleaning missing values with mean and median, and applying one-hot encoding using pandas get_dummies.

Standardize and normalize selected features with pandas and scikit-learn, verify transformations with pre- and post-operation summaries, then filter loyal employees by years at company and analyze attrition trends for retention.

Engage in hands-on data wrangling and analysis with Walmart sales data: perform missing value imputations, feature engineering, standardization and normalization of sales, one-hot encode store location, visualize sales by department.

Master data wrangling and feature engineering on Walmart sales data by handling missing values, applying imputation, and performing standardization, normalization, and one-hot encoding.

Perform data wrangling and analysis with imputation, missing value handling, standardization, normalization, and one-hot encoding, then engineer features like revenue and date components using Python and pandas.

Explore data visualization essentials with Python libraries and generative AI, including ChatGPT, which writes Python code. See code analysis and apply visualizations to stock, cancer, and Walmart datasets.

Explore financial planning and analysis and learn to visualize financial data with Python libraries Matplotlib, Seaborn, and Plotly, including interactive charts, heat maps, and scatter plots.

Discover data visualization types, including scatter plots, bubble charts, bar charts, line charts, and pie charts. Explore histograms and box plots for distributions, mean, median, standard deviation, and outliers.

Master key stock data terms, including open, high, low, close, volume, and adjusted close, and compute daily returns with pandas percentage_change for visual stock analysis and missing value handling (NaN).

Visualize Amazon stock data in pandas, compute daily returns, and create line, volume, pie, and candlestick charts using seaborn, matplotlib, and Plotly while noting no-code trends and future GPT development.

Explore practical techniques to visualize a multi-stock portfolio, compute daily returns, plot histograms, and build a correlation heat map to assess diversification and sector linkages.

Visualize breast cancer data using seaborn and matplotlib to explore features like mean area and mean smoothness, plot scatter plots and heatmaps, and analyze class balance.

Visualize Walmart store data to map revenue trends over weeks, compare revenues with visits, examine weather impact, and identify turnover extremes, using generative AI to create diverse visualizations.

Explore Walmart data visualization with generative ai by building line charts, scatter plots, and bar charts from prompts to reveal revenue trends, correlations with visits, weather effects, and turnover insights.

Generate synthetic Walmart store data with Claude by Anthropic for ten stores in top US cities, including weather-consistent weekly revenue and visits, then save to data.csv.

Master basics of machine learning for financial prediction, including supervised vs unsupervised learning, regression, and data splits. Apply decision trees, random forests, and generative AI to forecast revenues and expenses.

Explore the fundamentals of machine learning, including data-driven models, neural networks, and learning algorithms, and compare supervised, unsupervised, reinforcement learning, deep learning, and generative ai.

Identify labeled and unlabeled data and apply unsupervised learning to unlabeled inputs. Assess data quality—accuracy, completeness, consistency, relevance, timeliness, bias, and accessibility—and explore sources like Kaggle, AWS, and UCI.

Learn the fundamentals of regression, from simple to multiple linear regression, and use least squares to fit the best-fit line and predict outcomes like house prices.

Review regression fundamentals, focusing on simple and multiple linear regression, and practice drawing the given straight-line equations. Solutions will be shown in the next lecture.

Solve a practice opportunity to learn how line equations reveal slope and y-intercept; compare positive and negative slopes, the origin, and their impact on y values in regression fundamentals.

Learn to evaluate regression models using metrics like MAE, MSE, RMSE, MAPE, and R-squared, interpreting residuals, predictions, and explained variance.

Split data into training and testing sets (80/20) to train and evaluate on unseen data, and add a validation set for hyperparameter tuning to prevent overfitting and data leakage.

Explore the intuition of decision trees for classification and regression, and how random forests ensemble multiple trees with majority voting to improve predictions in a generative ai context.

See a live genAI demo that builds a regression model using Apple balance sheets, income statements, and cash flow data to forecast the next quarter earnings per share.

Encode fiscal period with one-hot encoding using pandas get_dummies, remove the column, split data with scikit-learn, train a random forest to forecast EPs, and evaluate with rmse, mse, and mae.

Predict Walmart's quarterly revenues and earnings per share using linear regression with features like store counts, housing starts, interest rates, consumer confidence, and evaluate with MSE and R-squared.

Explore Walmart quarterly data and prepare two regression models to predict revenue and earnings per share, using store count, housing starts, and macroeconomic features, with scaling and train-test split.

Train and evaluate linear regression models to predict Walmart quarterly revenue and earnings per share, and assess performance with mean squared error, root mean squared error, and R-squared.

Learn to use generative AI to benchmark your company against competitors, analyze marketing and financials, and build custom GPTs with embeddings and fine-tuning for competitive analysis.

Use generative AI to perform competitive analysis by extracting and summarizing data from sources, analyze marketing, pricing, and customer feedback, and automate SWOT-based insights.

Explore how to create and deploy custom GPTs—tailored, shareable AI agents that ingest your data, enable code interpreter and browsing, and align with brand voice and compliance.

Explore how embeddings convert words into dense vectors, place them in a semantic feature space, and reveal relationships across dimensions like age, gender, and royalty for language models.

Fine-tuning large language models adapts a pre-trained base model to a specific task or domain using task-specific data, creating a specialized model with updated weights through supervised training.

Learn how low rank adaptation (Lora) reduces fine-tuning compute and memory by decomposing weight updates into two small matrices A and B for language models.

Discover how retrieval augmented generation blends a retriever with augmented prompts and external knowledge sources to ground outputs from language models, boosting accuracy and up-to-date information.

Design and develop a custom GPT to perform competitor analysis using 10-K reports for Home Depot and Lowe's. Extract financial metrics, run SWOT and marketing analyses, and provide actionable recommendations.

Create a custom GPT for competitor analysis using 10-K reports, extract financial metrics, compare Lowe's with peers, perform a SWOT, and enable capabilities like code interpreter and web browsing.

Design a competitive analysis using genai prompts to compare Lowe's and Home Depot 2023 10-K data, grounding results and visualizing net sales, net earnings, and growth rates with bar charts.

Perform a data-driven SWOT analysis of Lowe's and Home Depot using their latest 10-Ks with a competitive analysis GPT; compare strengths, weaknesses, opportunities, threats across omnichannel, pro customers, and sustainability.

Analyze Lowe's vs. Home Depot using GenAI for debt-to-equity, ROI, and gross margin. Explore growth, supply chain, innovation, and risk factors from 10-K analyses.

Learn to use generative AI for competitor analysis by comparing Lo and Target, reviewing the 10-K, and assessing business model, risks, MD&A trends, revenue, and marketing metrics, plus drafting prompts.

Explore competitor analysis with GenAI by comparing Lowe's and Target using 10-K data, visualizations, and a SWOT, focusing on net sales, earnings, and growth.