Developing Generative AI Applications on AWS (Bedrock)

Python is a high-level, interpreted language with readable syntax for beginners. Use its libraries for web development, data analysis, and machine learning, including Django, Flask, Pandas, and TensorFlow.

Explore Python data types such as integers, floats, strings, and booleans, learn type conversion, and master arithmetic, comparison, and logical operators for hands-on labs.

Master Python lists and tuples, including indexing, slicing, and operations like append and remove, plus list comprehensions and tuple immutability for constants and data protection.

Master dictionaries and file handling in Python by using key-value pairs, accessing data with keys, and performing updates and iterations; explore inbuilt and user-defined functions, variables, and practice with Colab.

Learn to implement arithmetic and comparison operators in Python using Colab, define variables, print results, and document code with text cells across addition, subtraction, multiplication, and division.

Master Python comparison operators, including equals, not equals, greater than, less than, and greater than or equal to and less than or equal to forms, with print examples and assignments.

Explore assignment answers for Python operators using x, y, and z to print comparisons like >, <=, ==, and !=, revealing true or false results.

Demonstrate python data types by declaring integer, float, string, and complex variables, printing their values and types, and showing use cases like phone numbers, prices, and text.

Explore Python strings, including single-line and multi-line formats, escape sequences, and slicing, then apply a hands-on assignment demonstrating int, float, and string data types.

Explore a Python data types assignment demo for the AWS Bedrock course by defining name as a string, score as a float, and age as an integer.

Explore Python lists with a hands-on e-commerce example, creating a products list, indexing specific items, and modifying the list with append, delete, and insert operations.

Explore Python lists in this demo: extend and append operations, slicing an inventory, and completing assignments to add a charger and accessories such as a screen protector and phone case.

Learn how to update a Python list using append to add a charger and extend to merge items like screen protector and phone case, then print the updated list.

Explore Python tuples and lists, comparing mutability, performance, and memory efficiency, and learn use cases where tuples are read-only and fast, while lists stay mutable, with append not supported.

Demonstrate Python tuples by deleting and inserting items, extending with plus, slicing, and building an inventory to compare tuple behavior with lists in an e-commerce scenario.

Explore Python dictionaries through a hands-on inventory example, learning key-value pairs, adding and updating items, calculating total inventory value with a prices dictionary, and identifying low-stock items using a threshold.

Learn to manage inventory with python dictionaries by removing zero-stock items, merging inventories with summed quantities, and filtering by categories like apparel, footwear, and accessories.

Review the Python dictionary inventory by filtering footwear items with for loop and printing each product and its quantity, then update jeans and sneakers.

Master Python file handling for machine learning problems by creating, writing, reading, and appending text files and using the CSV module to write and read CSV data.

Learn to download data from URLs, handle CSV and text responses, and process logs by extracting errors into a dedicated error file using Python's requests, csv, and StringIO.

Explore Python inbuilt functions like len, max, min, and range, and define your own function. Apply input, print, and basic ecommerce calculations in hands-on Colab examples.

Define and call Python functions to calculate total price from price and quantity, display product details, sum cart item prices, and compute tax inclusive totals.

Demonstrates creating and updating Python variables in a Colab notebook, printing values, overwriting assignments, and identifying data types (int, float, string) with type().

Demonstrates Python variables by concatenating first name and last name, performing arithmetic, and combining text with numbers to calculate total cost. Includes an ecommerce subtotal and 10% discount calculation.

Watch a Python variables assignment demo that defines item_price and quantity, computes subtotal, applies a 10% discount, and prints subtotal, discount amount, and final price.

Explore machine learning fundamentals—supervised, unsupervised, and reinforcement learning—along with data preparation, TensorFlow workflows, and model evaluation, including a cat-vs-dog classification example.

Explore how machine learning powers voice assistants, autonomous vehicles, personalized recommendations, and generative AI, highlighting efficiency at scale, pattern recognition, predictive insights, continuous improvement, and enhanced customer experiences.

Explore supervised learning with labeled data, covering regression and classification tasks like house prices and spam detection, plus image classification, using training and testing data to improve accuracy.

Unsupervised learning trains on unlabeled data to uncover hidden patterns via clustering, customer segmentation, and association for promo offers. TensorFlow demonstrates pattern discovery, grouping cars and bikes without labeled outcomes.

Observe how reinforcement learning trains an autonomous vehicle to act in traffic through trial and error, using exploration, rewards, and a developing policy to maximize safety and efficiency.

Explore the differences between supervised, unsupervised, and reinforcement learning with real-world examples such as spam classification, sentiment analysis, image classification, and autonomous vehicles, supported by TensorFlow, PyTorch, and scikit-learn.

Explore how neural networks mimic the human brain to recognize patterns, powering deep learning with input, hidden, and output layers processing unstructured and unlabeled data.

Explore real-world deep learning applications such as speech recognition, image recognition, and natural language processing, with examples from voice search, photo tagging, and ChatGPT-style text generation.

Explore the decision tree model, a versatile and intuitive tree-structured approach for classification (and regression), with real-world uses in banking risk analysis, e-commerce segmentation, and medical diagnosis apps.

Learn linear regression as a supervised learning method, from simple linear regression with one independent variable to complex models with multiple variables, for predictive analysis of outcomes like house prices.

CNNs are deep neural networks that automatically learn features from visual imagery, enabling image and video recognition, facial recognition, and medical imaging applications with translation invariance.

RNNs, or recurrent neural networks, recognize patterns in sequences like text, handwriting, and time-series data. They enable real-time translation, sentiment analysis, and weather forecasting using TensorFlow, Keras, PyTorch, and MXNet.

Outline four ML algorithms—linear regression, decision trees, CNNs, and RNNs—and their practical use cases in pricing, loans, imaging, recognition, translation, and text generation.

Improve model predictions by prioritizing data preparation and quality, and applying cleaning techniques such as handling missing values, normalization, feature encoding, and train-test splits.

Explore ethics in machine learning for generative AI on AWS Bedrock, focusing on mitigating bias, protecting privacy and security, and ensuring transparency and accountability in ML models.

Discover the future of machine learning as AI and ML integrate into everyday devices, autonomous vehicles, and MS Office, while ethical and explainable AI shape jobs and impact every industry.

Explore numpy basics by creating and reshaping arrays, using zeros and ones, generating ranges with arange, performing arithmetic, indexing and slicing, and computing min, max, and sum.

Explore pandas fundamentals by building a data frame from a Python list, defining columns, saving and loading CSV files, and adding a discount price column for analysis.

Demonstrate creating a numpy array and converting it to a pandas data frame, then save and read CSV files to show the connection between numpy and pandas for machine learning.

Build a supervised image classifier with TensorFlow and Keras by preparing cat and dog data, splitting it into training and validation, and training a convolutional neural network.

Train model with image data generator, resizing images to 150 by 150, using a 20-image batch for training and validation on cats and dogs across 15 epochs to improve accuracy.

Save the trained model as an .h5 file. Load the model with TensorFlow Keras and prepare a test image by converting it to an array and adding a batch dimension.

Save the cat dog model with a .h5 extension, load it in TensorFlow Keras, and preprocess a 150x150 test image to an array with a batch dimension in Google Colab.

Explore unsupervised learning by building a synthetic e-commerce dataset with NumPy and pandas in Colab, using a reproducible seed to generate customer names, purchase amounts, categories, and browsing data.

Explore unsupervised learning with pandas and numpy to build a 500-row e-commerce data frame, standardize purchase amounts, apply k-means clustering into three spend groups, and analyze category and city sales.

this demo shows single-variable linear regression to predict insurance premium from age, using sklearn linear regression, numpy arrays, and optional csv data with pandas and seaborn for visualization.

Demonstrates multi-variable linear regression for insurance premium prediction using age, height, and weight. Shows data cleaning with mean imputation and model fitting to compare multi-variable to single-variable approaches.

Explore what generative AI is and its real-world applications across industries, with live examples, tools, libraries, ethics, security, and how to manage AI projects now and in the future.

Define generative AI as a class of artificial intelligence that creates text, images, music, and video from data by mimicking learned material, tracing evolution from 1950s foundations to neural networks.

Learn to generate a complete video from a single prompt using a freeware tool, with background music, AI voices, and fonts for platforms like YouTube, Facebook, and Instagram ads.

Explore applications of generative AI across content creation, personalized education, healthcare, gaming, fashion, advertising, finance, legal drafting, robotics, urban planning, and customer service.

Explore neural networks, including CNN and RNN, GANs, VAEs, diffusion, autoregressive and transformer models. See how reinforcement learning and fine-tuning tailor pre-trained models for image, text, and music generation.

Demonstrates generating PowerPoint slides with a single prompt using a Gen AI app, including outline, cards, images, and animations, and exporting to PDF or PowerPoint.

Assess ethics, privacy, and legal aspects of generative AI, emphasizing fairness, bias prevention, accountability, and transparency, with GDPR and CCPA considerations and risks like surveillance and deepfakes.

Define the scope and objectives for gen AI projects across industries, prioritize milestones such as multilingual chatbot support and robotic surgery, and monitor with dashboards, agile methods, and data governance.

Expose common GenAI security threats by detailing adversarial, vision, poisoning, privacy, and backdoor attacks, and stress clean training data to build robust security features.

Explore the future of AI, highlighting automation-driven job displacement, upskilling and AI literacy, ethical considerations, and evolving ML advances, compute power, and big data.

Learn to generate ai voices for professional voiceovers using a versatile tool that supports multiple languages and male or female voices, with options to customize accent and download audio.

GenAI in finance enables anomaly detection for fraud and network analysis to uncover fraud networks, while automating reporting and cash-flow forecasting with process automation and multilingual chatbots for 24/7 support.

Explore how gen ai enables personalized product recommendations and data-driven marketing campaigns. Identify sales-ready leads with predictive scoring, automated outreach, real-time optimization, and sentiment analysis for trends.

Automates applicant screening, resume parsing, and personalized candidate assessments with GenAI in HR management. Predicts retention risks and delivers data-driven, unbiased performance feedback through continuous monitoring and automated performance management.

Demonstrates generating AI avatar videos by cloning yourself with AI Studio, scripting with ChatGPT, and rendering AI news avatars with voiceovers for a cloud-enabled channel.

Explore generative AI in healthcare for personalized medicine and early diagnosis. Leverage computer vision for medical image analysis and robotic surgery, with ethical considerations for responsible use.

Prompt engineering is the art of writing effective instructions for AI models to generate desired outputs. Iterative, data-driven refinement boosts quality, productivity, and creativity by defining tasks, constraints, and context.

Master zero-shot, few-shot, and chain-of-thought prompting techniques to shape outputs in generative AI applications on AWS Bedrock, with context-based, iterative, and prompt optimization strategies.

Trace how a prompt becomes output through input encoding into tokens, model initialization, context processing, and token generation with probability estimation and text assembly.

Explore prompting with ChatGPT 3.5, using universal prompts to shape outcomes, and apply examples like content generation, constraints, translations, video scripts, and business-planning prompts.

Explore ChatGPT-4 capabilities, including web search, attachments, PDF summarization, code generation for websites, CSV data creation, and crafting SEO-ready social posts with images.

Explore prompting with anthropic Claude cloud, sign up, and run practical prompts that demonstrate Python coding tasks, code optimization, debugging, and building simple web apps, with latest data access.

Explore prompting with Google Gemini to write programs, translate text, and draft blogs, while comparing free and advanced versions and noting its Gmail and work docs integration.

Explore natural language processing (NLP), including natural language understanding (NLU) and natural language generation (NLG), applications like sentiment analysis, and the NLP pipeline from data acquisition to model evaluation.

Explore natural language processing applications such as language detection, translation, smart email replies, voice assistants, spell check, and multilingual customer service bots.

Trace NLP's evolution from heuristic rule-based methods to machine learning and deep learning, using child-like analogies, and explore rule-based parsing, regex, and expert systems as early approaches.

Explore NLP methods from supervised and unsupervised learning to deep learning with Naive Bayes, SVMs, CNNs, RNNs, LSTMs, and transformers for sentiment analysis and text classification.

Explore key NLP challenges, including ambiguity, contextual meanings, irony and tone, spelling errors, dialects, slang, and rapid contextual evolution, and their impact on understanding and generation.

Explore text classification and sentiment analysis, information retrieval, and natural language queries using ML models like Naive Bayes, SVM, Random Forest, and transformers.

Master language detection and translation using SMT and NMT, and build multilingual pipelines with real-time, context-aware translation. Explore conversational agents, NLU, NLG, and knowledge graphs for robust NLP solutions.

Explore text summarization and topic modeling with lda, nmf, and hdp, and apply text generation and speech-to-text to news, academia, and conversational ai using deep learning.

Learn the six stages of building an NLP pipeline—from text acquisition and pre-processing to feature extraction, application integration, and model evaluation—through practical sourcing and data techniques using real-world ecommerce reviews.

Explore an end-to-end nlp pipeline for sentiment analysis of amazon reviews, from text extraction with web scraping to cleaning, feature extraction, and classification with bert or svm.

Build an email spam filter in Python using Google Colab, pandas, NLTK, and a count vectorizer to classify emails as spam or not spam, with setup and preprocessing.

Demonstrate text summarization using NLTK and the LSA summarizer in a Colab workflow, including tokenization and stopword removal, to convert long content into a concise three-line summary.

Learn NLP data pre-processing to cleanse data from web scraping, APIs, and databases by applying lowercasing, html tag removal, url and punctuation removal, emoji stripping, spelling corrections, and stopword removal.

Explore large language models, their industry use, programmatic access, fine-tuning with Hugging Face, and multimodal language model concepts, while discussing advantages, limitations, ethics, and the future of lm products.

Explore large language models built on transformer architectures and trained with unsupervised pre-training and fine-tuning. Understand their applications and limitations, including bias and privacy, across healthcare, finance, and other industries.

Explore how LMS enables text-to-image, image-to-image, and text-to-video generation, plus music and translations. Discover applications in sentiment analysis, code generation, creative writing, and chatbots with voice interfaces.

Understand the difference between custom and fine-tuned language models by defining objectives, preparing data, selecting architectures, and training and evaluating models with pre-trained options.

Explore multimodal lms and llms that process and integrate text, images, and audio to generate end-to-end solutions for healthcare, education, and interactive assistants.

Learn to access OpenAI programmatically with Python by installing OpenAI and requests, configuring api keys, and using the chat completion endpoint with gpt-3.5 turbo to build generative ai apps.

Explore OpenAI LLM NLP tasks like text summarization, translation to Spanish, and text completion. Build a simple chatbot using prompts with adjustable temperature and Python code.

Learn to access Anthropic Claude programmatically with Python in Colab, obtain API keys, set up the client, and build basic prompts and chat interactions.

Demonstrates Claude LLM for NLP tasks on cloud platforms, including text summarization, translating to French, and building a cloud-enabled chatbot using API keys and the client message create method.

Fine-tune distilbert for text classification using hugging face on Colab. Build a dataset of harmful, spam, and generic messages and train with the trainer to produce a fine-tuned distilbert model.

Develop an OpenAI chatbot with a Python Flask web interface, guided setup, and error fixing, including Colab demos and API key integration.

Upload a PDF to the chatbot, then ask questions to summarize or extract information. Build with Python, Flask, and OpenAI, using pdf plumber and GPT-3.5 Turbo.

Build a text-based chat bot that queries OpenAI and converts replies to speech for playback in the browser.

Set up a web ai chatbot environment by provisioning a VM with two CPUs, four gb ram, and 40 gb disk on any cloud, then install python 3 and pip.

Demonstrates deploying a web chatbot on an AWS EC2 Linux VM, using Colab for notebook steps, and setting up Docker, Python, with Fei data PG vector and a streamlit app.

Provide an OpenAI API key, connect a target website, and let the chatbot crawl and learn from the knowledge base using a Postgres vector DB for accurate answers.

Build a telegram openai bot with a ChatGPT integration using Telegram and OpenAI APIs, following Google Colab steps and token setup to enable chat prompts and responses.

Explore hugging face and its transformer library for NLP, featuring pre-trained models like GPT-2 for text classification, sentiment analysis, and entity recognition; learn fine-tuning with datasets and setup with pip.

Explore how Hugging Face transformers power NLP tasks—text generation, sentiment analysis, translation, summarization, question answering, automatic speech recognition, and named entity recognition—using the simple pipeline interface.