A2A Protocols & APIs-Build Real-World AI Systems with Python

Explore APIs and REST principles, build endpoints with fast api using get, post, put, and delete methods, and test with interactive swagger docs.

Build a user management API using FastAPI and Pydantic to perform CRUD on an in-memory database, with routes for get all, get by id, create, update, and delete.

Learn to test a FastAPI /users API with Postman, performing get, post, put, and delete operations on a local in-memory database and handling 200, 201, and 404 responses.

Develop a fast api app that uses a Hugging Face sentiment analysis model to classify text as positive or negative with a confidence score via the feedback and excel endpoints.

Automate the handling of customer feedback with sentiment analysis to automatically escalate negative messages into CRM tickets and alert teams via Slack, improving response speed and preserving brand trust.

Build a sentiment analysis API with fast API and a hugging face model to classify text as positive, negative, or neutral, enabling Excel batch processing, CRM tickets, and Slack alerts.

Automate escalation of negative feedback by auto creating Zoho Desk tickets and sending Slack alerts, triggered by sentiment analysis via the Hugging Face model in analyze and batch analyze endpoints.

Enhance the sentiment analysis API with robust error handling, tokenization and truncation to 512 tokens, and reliable alerts via Zoho Desk tickets and Slack, preventing batch crashes and data loss.

Build an AI email classifier that fetches emails via Gmail or SendGrid APIs, uses Bard or Spacy for classification into support, sales, or spam, and routes to Zendesk or Salesforce.

Learn to authenticate with Gmail via the Google API, fetch and decode emails (HTML to plain text when needed), and print them in a clean format, with optional SendGrid integration.

Classify emails with a Bard-based zero-shot model, labeling topics like support, sales, HR, finance, payment, and technical; integrate with Gmail API data and report confidence scores.

Route classified emails to the correct endpoints by forwarding support to a dummy Zendesk API and sales to a dummy Salesforce API, then via SMTP with TLS to real addresses.

Build a chatbot using Dialogflow and FastAPI to detect intents and extract city data. Call real APIs, such as Openweathermap, to respond with live weather and city-specific details.

Set up a FastAPI backend to receive Dialogflow intents via a webhook, extract the chatbot intent and city, and fetch weather data from OpenWeatherMap using a dot env API key.

Create an intent to trigger an external API using Google Dialogflow. Train phrases and map a location parameter to a webhook that calls the Openweathermap API.

Build a smart chatbot flow using Dialogflow for intent detection and parameters, with a fast API backend calling the Openweathermap API service to deliver real-time weather responses via webhook.

Discover agentic AI, where proactive agents pursue goals autonomously using perception, reasoning, and action with memory. Explore API integration, multi-agent systems, safety guardrails, and impacts finance, logistics, education, and healthcare.

Build and test an ai agent that selects actions in event-driven workflows, using weather, email, and scheduling tools called by a Gemini model via structured json prompts.

Queue AI tasks with Redis to enable asynchronous, event-driven workflows powered by an agent. Process queued tasks from Redis with a background worker, enabling scalable, fault-tolerant tool execution and logging.

Learn to build real-world ai systems with rest apis, Flask, and FastAPI, sentiment analysis, crm automation, and event driven workflows, plus best practices for security, reliability, and collaboration.

Explore three directions: LangChain for building with large language models, memory, and API connections; MLOps for production model versioning and deployment; and GPT-based agents for planning, reasoning, and action.