Theme1

100 Days of GenAI Challenge: LinkedIn Showcase Edition

Theme:

From Python Basics to Building Agentic AI Workflows

Phase 1: Foundation Setup (Days 1–10)

  • Day 1: Install Miniconda/Anaconda

  • Day 2: Create a Conda Environment and install Python 3.11

  • Day 3: Basics of Python: Variables, Data Types

  • Day 4: Basics of Python: Functions and Loops

  • Day 5: Basics of Python: Object-Oriented Programming (OOP)

  • Day 6: Git and GitHub Setup – Create your first Repository

  • Day 7: Jupyter Notebook Setup

  • Day 8: Learn basic Markdown (for documentation)

  • Day 9: Install VSCode + Python Extension

  • Day 10: Post your environment setup screenshots and a simple Hello World Notebook on LinkedIn

Phase 2: Machine Learning Basics (Days 11–20)

  • Day 11: Install scikit-learn and pandas

  • Day 12: Build your first Linear Regression Model

  • Day 13: Classification with Decision Trees

  • Day 14: Data Preprocessing Basics (missing values, scaling)

  • Day 15: Train-Test Split + Cross-Validation Basics

  • Day 16: Introduction to Huggingface Datasets

  • Day 17: Simple Sentiment Analysis Dataset (IMDB)

  • Day 18: Train a basic Random Forest on a dataset

  • Day 19: Share your ML Model performance on LinkedIn

  • Day 20: Refactor your code into clean scripts (Python modules)

Phase 3: GenAI Core Skills (Days 21–40)

  • Day 21: Introduction to LLMs (basic theory)

  • Day 22: Install OpenAI API, complete your first text generation

  • Day 23: Temperature, Top_p, Max Tokens explained

  • Day 24: Build a simple Prompt Engineering app (input box -> output)

  • Day 25: Few-shot Prompting Examples

  • Day 26: Chain of Thought (CoT) Prompting

  • Day 27: Zero-shot, One-shot, Few-shot case studies

  • Day 28: Introduction to Huggingface Transformers

  • Day 29: Load a pretrained LLM like GPT2 locally

  • Day 30: Build a simple chatbot using Huggingface Pipeline

  • Day 31: Introduction to Embeddings

  • Day 32: Generate sentence embeddings with OpenAI API

  • Day 33: Install FAISS locally

  • Day 34: Create your first RAG (Retrieval Augmented Generation)

  • Day 35: Document and post your RAG mini-project

  • Day 36: Load PDF and generate embeddings (using unstructured/llamaindex)

  • Day 37: Compare local LLM vs OpenAI LLM (output quality)

  • Day 38: Try Claude or Gemini APIs

  • Day 39: Create an evaluation sheet comparing models

  • Day 40: Write a LinkedIn post: "My First Month in GenAI: Learnings"

Phase 4: Advanced GenAI Skills (Days 41–70)

  • Day 41: LangChain Basics

  • Day 42: Build a basic Chain (prompt -> LLM -> output)

  • Day 43: Multi-step Chains (sequential chains)

  • Day 44: Memory in LangChain

  • Day 45: Custom Tools in LangChain

  • Day 46: LangChain Agents Introduction

  • Day 47: Tool Usage with LLMs (like Search, Calculator)

  • Day 48: Build your first LangChain Agent

  • Day 49: Learn what VectorDBs are: FAISS, Chroma, Qdrant

  • Day 50: Connect LangChain with FAISS Vectorstore

  • Day 51: Building a Personal Document QA Bot

  • Day 52: Autonomous Agents Introduction (BabyAGI, AutoGPT)

  • Day 53: Install and run a local BabyAGI variant

  • Day 54: Understand Task Decomposition in Agents

  • Day 55: Self-Refinement (Reflection) Concepts

  • Day 56: Prompt Engineering for Agents

  • Day 57: Experiment with goal-driven Agents

  • Day 58: Understand Function Calling in OpenAI API

  • Day 59: Create a Function Calling Demo (dynamic JSON outputs)

  • Day 60: LinkedIn Post: "How I built my first GenAI Agent"

Phase 5: Professional Deployment Skills (Days 71–90)

  • Day 71: Docker Introduction for AI apps

  • Day 72: Dockerize a basic RAG App

  • Day 73: Introduction to FastAPI

  • Day 74: Create a simple GenAI API using FastAPI

  • Day 75: Serve your RAG bot as a FastAPI app

  • Day 76: Prefect Basics (flows and tasks)

  • Day 77: Automate a small GenAI job with Prefect

  • Day 78: Understanding Work Pools and Deployments

  • Day 79: Deployment to Render/EC2/any free tier

  • Day 80: LinkedIn Post: “How I Deployed My GenAI App”

Phase 6: Agentic Workflows & MCP, A2A (Days 91–100)

  • Day 91: Introduction to MCP (Model Context Protocol)

  • Day 92: Read and understand ModelContextProtocol SDK

  • Day 93: Build a simple MCP Server (Stdio Transport)

  • Day 94: Write tools/functions for your own MCP Agent

  • Day 95: Understand A2A (Agent-to-Agent) communication

  • Day 96: Build a simple 2-agent system (one writer, one editor)

  • Day 97: Add memory context to agents

  • Day 98: Simulate a Multi-Agent Workflow: Researcher -> Writer -> Editor -> Publisher

  • Day 99: Polish your MCP Project + GitHub README

  • Day 100: Final LinkedIn Post:

"From Python to Autonomous Agent Workflows — My 100 Days GenAI Journey"

LinkedIn Post Template Suggestions:

  • Weekly summary posts (Day 7, Day 14, Day 21, etc.)

  • Mini-project showcases with screenshots

  • #100DaysOfGenAIChallenge hashtag

  • Tagging new connections, peers, and GenAI leaders

PreviousGenAI 100 Days ChallengeNextTheme2

Last updated