10 Gen AI Projects in 10 Days: From Zero to AI Expert

Set up a local generative ai text generator project with the Gwen model via Ollama, install Ollama, and run a localhost server for text generation.

Learn to load the Gwen model and generate text with the gen text function using prompt, temperature, and limit, with a post to the Olama server and try-except error handling.

Define a main function and loop to capture user prompts, generate text with a gen text function, and exit when the user types exit. Validate prompts and print the output.

Set up a genai project for an ai email generator using llama 3, install ollama, configure a local server on localhost:11434, and import the llama 3 model to generate outputs.

Create an email generation function that uses subject, recipients, purpose, and tone as inputs. Build a structured prompt for the model and fetch responses from a local lama3 server.

build a Python main function that collects subject, recipient, tone, and purpose to generate and print an email with a gen email function.

Set up a gen AI image generation project using the Stable Diffusion model, handling GPU or CPU execution, and installing diffusers, hugging face hub, and torch in Python.

Demonstrates loading a pre-trained stable diffusion model, configuring data type and device, enabling attention, prompting the model to generate and save an image (sample.png) with an apple example.

Set up a language translation project by creating an app.py in a new folder, installing Olama and the GWEN model, configuring localhost:11434, and preparing to build translation functions.

Learn to implement a translate function that builds a model prompt with the target language and input text, posts to a local model, and returns the translated output.

Learn to read a pdf from a file path using a pdf reader package, extract text from each page, and assemble it into a final text output.

Analyze resumes against job descriptions to predict ATS scores using a structured prompt, extracting keywords, strengths, weaknesses, and actionable suggestions to improve resume performance.

Create a main function to read a pdf, input a job description, generate an ATS score with lama3, and display feedback on strengths and missing keywords.

Install and configure Ollama, download the lama3 model, and run a localhost 11434 server to build a YouTube video summarizer that processes transcripts via the transcript API.

Learn to extract the YouTube video ID from URLs using a Python function, handling URL formats using slashes and the question mark as split points, and prepare for transcript extraction.

Fetch the transcript of a YouTube video using its ID via the YouTube transcript API, join the sentences into text with a for loop, and return the summarization.

Explore loading a model by fetching website content with the beautiful soap package, removing script and style tags, extracting text, capping at 3000 characters, and handling errors.

Explore prompt engineering to summarize website content using a sum website function and prompts for the model, returning a JSON summary.

Build a console app that prompts for a website URL, uses a main function to fetch and clean content, then summarizes it with the Mistral model and prints the result.

Set up a drag pdf chatbot project using the Gwen model to summarize a sample pdf, introduce rag for source content, and configure a python app with olama.

Create a function to split text into 500-character chunks, store each chunk in a list, loop over the text, append chunks, and return the output.

Learn to convert text to embeddings by building a get_embeddings_of_text function, and call the nomic embed text model via the olama server with a text prompt.

Create a vector database to store embeddings from chunks and enable fast search with Faiss by building a l2 index, converting embeddings to a NumPy float32 array, and adding them.

Create a search function that converts a user query to embeddings, searches a FAISS index for the top 3 similar chunks by embedding similarity, and returns the results.

learn how to craft a prompt with context and question, structure a prompt block, and request a model response via a server to obtain a json output.

Learn to build a pdf chatbot using rag: load a pdf, split text into chunks, create a vector database, and answer queries via a Gwen llm.