Lifetime Dashboard Access

Live Doubt Clearing Sessions

Resume Discussion And Mock Interview

Paid Internship Opportunity In Research Project for Freshers

What is Generative AI?

History and evolution of AI and Generative AI

Use cases across industries

What are LLMs?

Transfermor Architecture and how they work

Popular LLMs: GPT, BERT, Claude, LLaMA

Training and fine-tuning LLMs

Understanding ablout Langchain & its Implementation

Introduction to prompt engineering

Types of prompts

Instruction-based prompts

Chain-of-thought prompts

Few-shot and zero-shot prompts

Advanced techniques for designing effective prompts

Evaluating prompt performance and bias

What are vector databases?

Role in modern AI applications

Why traditional databases fall short for similarity search

Differences in data storage and retrieval

Use cases for each

Indexing and search efficiency

High-dimensional vector space

Distance metrics: cosine similarity, dot product, Euclidean

Approximate Nearest Neighbor (ANN) search algorithms

Overview & Complete Implementation: FAISS, Pinecone, Weaviate, Qdrant, ChromaDB

Pros and cons of each Vector DB

Generating embeddings using OpenAI, HuggingFace, Cohere, etc.

Introduction to RAGs

RAG vs traditional LLM generation

How RAG works

Two-phase architecture: Retrieval + Generation

Overview of Dense Passage Retrieval (DPR)

Role of vector databases (e.g., FAISS, Weaviate, Pinecone)

Architecture of a RAG pipeline

Handling structured vs unstructured data

Use of embedding models (e.g., OpenAI, HuggingFace, SentenceTransformers)

Integration of LLMs with external data using RAG

Building a RAG-based system

Connecting LLMs to knowledge bases for real-time information generation

Introduction to AI agents

Single-agent vs. multi-agent systems

Understanding Agentic AI

Characteristics of Agentic AI

Real-world applications (e.g., customer support, autonomous systems)

Building intelligent agents using LLMs

Integrating with APIs

Context retention and action-based responses

Build a system where users upload PDF, DOCX, or image files

Extract content and allow users to ask questions in English, Hindi, or Marathi

Provide accurate, context-aware answers using a RAG architecture

Ensure OCR + QA performance on noisy scans

Create test cases with documents in all three languages

Understand how to apply RAG architecture to real-world document Q&A

Learn to build multilingual applications using translation + language detection

Gain hands-on experience with file parsing, OCR, embedding models, and vector search

Integrate LLMs with external tools and pipelines for domain-specific Q&A

Understand the core goals: transcribing spoken content, identifying multiple speakers, and summarizing long conversations.

Explore real-world applications such as podcasts, interviews, meetings, and customer support analytics.

Learn how to extract audio from YouTube videos and local video files

Measure and validate audio duration before processing.

Use and Develop open-source transcription model on local machines.

Compare speed, accuracy, and latency between APIs.

Building Speaker Diarization Module

Detect and differentiate between multiple speakers using WhisperX’s diarization module.

Assign speaker labels (e.g., Speaker 1, Speaker 2) to each word or sentence.

Align words and segments accurately with timecodes.

Understand benefits of character-level alignment for subtitle creation or searchability.

Structure output for easy integration into UIs, dashboards

Handle errors like missing audio, failed transcription, or model timeouts.

Track processing time and log detailed outputs for auditing.

Improve performance by caching models and batching requests.

Design ideas for deploying a Streamlit or React-based UI.

Learn how to integrate APIs from multiple content platforms ( Youtube, Medium, News Channel, Instagram )

Understanding about project Architecture

Multi-Platform Content Aggregation

Retrieves latest articles from specified authors

Understand AI-based summarization using LLMs

Content Processing & Generates concise and meaningful summaries

Maintains context and key information from original content

Digest Generation

Email Delivery System & Sends formatted markdown digests via email

Build a full-stack content pipeline (Fetch → Summarize → Format → Send)

Use Twilio or n8n + Twilio to handle incoming/outgoing voice calls

Use n8n as the central no-code automation platform to orchestrate workflows

Integrate Twilio (via n8n) to handle incoming and outgoing voice calls

Connect with Retell AI to power real-time conversation and generate human-like voice responses

Design conversational flow visually inside n8n using trigger, condition, and HTTP request nodes

Automate follow-up actions like sending SMS, saving to Google Sheets, or emailing call summaries

No coding required — fully drag-and-drop based configuration with reusable templates