GenAI Jumpstart (Full Course)

Begin your journey with Generative AI by exploring the possibilities it holds for your career, why this technology matters, and how it can give you an edge in the workplace.

Understanding how fear, apathy and confidence play a role in how we are using AI, where you fall on the AI Attitudes Curve and how we can shift our mindset to embrace GenAI with confidence.

Get to know your instructor (me, Kaitlyn!) – a seasoned entrepreneur and software engineer passionate about AI, with a decade of experience in building and scaling tech companies.

A look ahead at what you'll learn in this course to accelerate, augment and automate parts of your role – from AI jargon to practical implementations, you'll be equipped to navigate and lead in AI projects confidently.

Explore the four main components of generative AI applications: the AI model, hosting infrastructure, additional data, and prompting. Learn how these elements work together to create powerful AI systems, using analogies to help understand complex concepts.

Dive into the transformative potential of generative AI across various industries. Understand the core functionality of AI models, their limitations, and the vast array of applications from personalised experiences to automated customer support, highlighting the importance of embracing this technology.

Unpack the relationship between artificial intelligence, machine learning, deep learning, and generative AI. Discover how these concepts build upon each other, using culinary analogies to illustrate the increasing complexity and capabilities of AI systems.

Compare and contrast generative AI with traditional AI approaches. Learn about the accessibility, cost-effectiveness, and innovative potential of generative AI, while understanding its place in the broader AI landscape and its implications for businesses and entrepreneurs.

This section helps to demystify all the GenAI and AI jargon and acronyms you may hear in conversations or meetings so you can speak the same nerd-language.

Experience a humorous yet enlightening roleplay showcasing the overwhelming use of AI terminology in a professional context, highlighting the need for clarity in communication.

Embark on a comprehensive exploration of AI jargon with Raven Mad, the sock puppet, covering fundamental concepts, model types, architectures, and key technical terms with relatable analogies to aid understanding.

Access a multi-page cheatsheet for AI jargon, covering fundamental concepts, model types, architectures, and key technical terms with relatable analogies to aid understanding.

Explore the vast array of generative AI models, from proprietary to open-source, and understand their growth projections and diverse applications across industries.

Dive into leading proprietary AI models like ChatGPT, Claude, and Gemini, comparing their strengths, use cases, and unique features.

Discover the world of open-source AI models, their accessibility, and the potential for customisation and data security in various applications.

Examine top image generation models like Midjourney and Ideogram, understanding their capabilities and limitations in creating visual content.

Learn how to effectively choose AI models using resources like LMSYS, HuggingFace and Ollama, understanding benchmarks and user preferences in model performance.

Step-by-step guide on finding, downloading, and running unrestricted open-source models on your personal computer.

Technical demonstration on setting up and using open-source AI models locally, including creating a user-friendly interface.

Open-source models on your smart phone that can be used offline and for free.

Introduction to unified LLM interfaces that allow access to multiple paid AI models in one place, simplifying model comparison and usage.

Guidelines for determining when and how to effectively implement generative AI in your workflow, focusing on necessity and return on investment.

An exploration of the top GenAI-driven applications or applications that feature GenAI for increased productivity and desired outputs, as well as platforms that list applications that might be more relevant for your particular role or industry niche.

Learn how to optimise your Minimum Viable Product (MVP) in AI implementation by balancing crucial factors such as speed to market, budget constraints, potential challenges, time investment, and data security concerns. Discover strategies to create an effective AI solution that meets your immediate needs while setting the stage for future scalability and success.

Understand the range of AI implementation options, from simple user interfaces to complex custom solutions, and how to choose the right approach for your specific needs and technical capabilities, demonstrated in a "Holy Grail" graph of GenAI options.

Understand the range of AI implementation options, from simple user interfaces to complex custom solutions, and how to choose the right approach for your specific needs and technical capabilities.

Examine real-world AI application scenarios across various industries, understanding how different implementation strategies apply to specific business needs and data requirements.

Delve into the critical issue of data security in AI applications, exploring common risks, incident rates, and strategies to protect sensitive information when using generative AI models.

Learn the key components of an effective AI usage policy for organisations, covering data sharing, compliance, and the balance between work and personal AI use.

Explore the ethical and safety aspects of different AI models, understanding how to evaluate and choose models based on their data handling practices and security measures.

Learn the basics of effective prompt engineering, including the importance of detailed prompts, the GIRDER methodology, and how to craft prompts for different types of tasks.

Dive into advanced prompting strategies such as zero-shot, one-shot, and few-shot learning, and understand how to use chain-of-thought prompting for complex tasks.

Learn the 3 types of prompts and discover techniques to enhance your prompts, including using specific writing styles, providing examples, and specifying output formats to get more accurate and tailored results + immediate "low hanging fruit" to enhance your prompts with just a few words.

Learn about meta-prompting - asking the AI about its own capabilities and needs - and how to iterate on prompts to refine outputs and achieve better results.

Explore resources for finding and using pre-made prompts, including online marketplaces and plugins, to save time and benefit from expert-crafted prompts.

Understand the benefits of conversational prompting and how to engage in effective back-and-forth interactions with AI to refine outputs and achieve desired results.

Learn how to train AI models to mimic your personal writing style or brand voice, using data examples and custom instructions for more authentic outputs.

Engage in targeted creativity exercises designed to enhance your prompt-crafting skills. Learn techniques to think outside the box, generate unique perspectives, and approach problems from different angles, all aimed at creating more innovative and effective prompts for AI interactions.

Sharpen your communication skills with practical exercises focused on clarity, concision, and context-setting in prompt writing. Practice articulating complex ideas simply, providing relevant details, and structuring information effectively to improve your ability to communicate with AI systems.

Explore the concept of multimodal AI, understanding different data types as inputs and outputs, and how this versatility enhances AI applications across various domains, including image, video, data analysis, audio and so forth.

Discover potential future developments in AI modalities, including sensory inputs and outputs, and how these advancements might shape future AI interactions and applications.

Learn to create data visualisations and charts using AI tools, understanding the process and best practices for generating visual representations of data.

Explore the creation of custom avatars using AI, understanding the applications and techniques for generating personalized digital representations for various uses.

Master the process of creating professional presentations using AI tools like Gamma, learning how to leverage AI for efficient and effective content creation.

Discover how to use AI to quickly generate professional PDFs, focusing on practical applications like invoice creation and report generation.

Learn advanced data analysis techniques using AI, understanding how to process large datasets and extract meaningful insights efficiently.

Explore the concept of autonomous AI agents, understanding their capabilities, applications, and how they can be utilised to perform complex tasks independently with an overview of Multi-Agent Systems and other 'multi-modalities'.

Delve into the world of AI phone agents, learning how to set up and use AI for handling calls, scheduling appointments, and managing other phone-based tasks in real-time and hands-off.

Discover how to create and deploy task-specific AI agents for various business processes, understanding their potential to automate and optimize workflows.

A step-by-step guide on setting up and running autonomous AI agents using tools like Crew AI and the OpenAI API, including practical demonstrations and coding examples.

Explore the concept of AI cloning for creating digital twins, understanding its applications, benefits, and potential ethical considerations.

Learn the process of AI voice cloning, including data requirements, tools like Eleven Labs, and techniques for achieving high-quality voice replications.

Master the creation of AI video avatars using tools like HeyGen, understanding the process, best practices, and applications for digital video representations.

Explore the intersection of AI and wearable technology, understanding how artificial intelligence is enhancing the capabilities of smartwatches, fitness trackers, and other wearable devices. Learn about real-time health monitoring, personalised insights, and predictive analytics that are revolutionising personal health management and daily productivity. Discover the latest trends in AI wearables and their potential impact on various industries, from healthcare to sports performance.

Introduction to four main techniques for customising the base generative AI model: prompt stuffing, fine-tuning, RAG, and embedding. Overview of their applications and benefits across different industries.

Detailed exploration of how different AI customisation techniques can be applied across various industries, including legal, accounting, marketing, software development, finance, and healthcare.

Emphasising the critical role of high-quality data in generative AI training, with guidance on collecting, preparing, and structuring data for optimal results.

Techniques for gathering and preparing training data, including using LinkedIn exports, web scraping, and leveraging external datasets for AI model customisation. Instructions on how to structure data in CSV format for AI training, with emphasis on cleaning and formatting data to match desired outputs.

Exploring the creation and use of synthetic data to augment limited datasets for AI training. Learn how to generate high-quality, artificial data that mimics real-world information, enabling more robust model training and overcoming data scarcity challenges in AI development.

Identifying and managing personally identifiable information (PII) in datasets, with strategies for data obfuscation to ensure privacy and compliance.

Exploring various sources for high-quality, clean external datasets, including Kaggle and other reputable platforms, to supplement your AI training data.

Detailed explanation of prompt stuffing, its benefits, limitations, and best practices for maximising context window usage in generative AI models.

Practical demonstration of prompt stuffing technique using an HR report generation example, showcasing how to leverage large context windows effectively.

Introduction to fine-tuning generative AI models, its applications, and available tools and platforms for implementing this technique.

Exploration of various resources and tools for fine-tuning, including pre-tuned models on Hugging Face and other platforms for efficient model customisation for your interest, a lot more technical!

Comprehensive overview of embedding in AI, its importance in understanding data relationships, and its applications in various industries.

Hands-on demonstration of implementing embeddings using OpenAI's API and Pinecone, with examples of both simple and complex data analysis.

Detailed explanation of Retrieval Augmented Generation (RAG), its advantages over other techniques, and various implementation methods including no-code solutions.

Step-by-step guide to setting up a local, private RAG system using open-source tools, demonstrating its application with a grant eligibility assistant example.

Exploring the power of function calling in generative AI models. Learn how to extend AI capabilities by defining custom functions, enabling models to perform specific tasks, access external data, or trigger actions. Understand the implementation, best practices, and potential applications of function calling to create more versatile and interactive AI systems. Use a no-code solution like Zapier to implement function calling with popular GenAI apps.

Exploring the challenge of AI hallucinations, their causes, and strategies to minimise them. Covers model selection, data quality, and post-processing techniques to improve output reliability.

Discussion on the varying reliability of different AI models, introducing the HHH leaderboard for hallucination rates, and highlighting upcoming models focused on reducing false information.

Detailed strategies for post-processing AI outputs, including prompt engineering for source verification and practical tips for fact-checking AI-generated information.

Addressing the gap between AI demonstrations and real-world performance, discussing cherry-picking in AI showcases, and managing expectations for AI capabilities.

Examination of ethical limitations and guardrails in AI models, comparing closed and open-source systems, and discussing the balance between functionality and responsible AI use.

Overview of the challenges facing generative AI in 2024, including context retention issues, generic outputs, and difficulties with text-to-image generation, highlighting areas for improvement in AI technology.

Discussion on the current state of AI adoption, emphasising that it's not too late to learn and implement AI. Explores the concept of 'practical AI' and its implications for the coming years.

Comprehensive overview of AI's progression from early text-based models to potential future developments, including advanced modalities, autonomous agents, and the long-term possibility of AGI, as well as how it sits in with the exponential growth of technology over the past 30 years.

Detailed explanation of Artificial General Intelligence (AGI), dispelling common misconceptions and outlining OpenAI's five-step journey towards achieving AGI, emphasising its complexity and future implications.

Encouragement to explore AI models beyond widely known options like ChatGPT, highlighting the benefits of less popular but potentially more effective models like Claude from Anthropic.

Guidance on how to stay updated with AI developments, including recommendations for following model updates, new modalities, and resources for continuous learning in the rapidly evolving AI landscape.

Final thoughts on the importance of experimenting with AI tools, emphasising ongoing improvements in AI models and touching on potential future applications, including controversial uses like digital clones of departed individuals.

A comprehensive summary of the course content, highlighting major topics covered including AI fundamentals, model customisation, ROI assessment, prompting skills, and strategies for staying ahead in the rapidly evolving field of generative AI.

Overview of additional resources and engagement options offered by the instructor, including in-person events, one-on-one consulting, workshops, and online communities for continued learning and networking in the field of generative AI.

Explanation of the course's reference materials and how the course structure and instructor qualifications allow it to count towards Continuing Professional Development (CPD) or Professional Development (PD) credits for participants.