🤖
Generative AI
Build the Future of Intelligent Applications
Salary Expectations (₹ INR, per annum)
Fresher
₹8 – ₹14 LPA
Mid-Level
₹15 – ₹28 LPA
Senior
₹30 – ₹60+ LPA
Detailed Learning Path
1
Deep Learning & Transformers
8–10 WeeksUnderstand the core architecture behind all modern LLMs — the Transformer model.
Key Topics to Cover
Deep Learning Theory — Neural Networks, backpropagation, and gradient descent.
NLP Pre-processing — Tokenization (e.g., BPE) and creating word/sentence Embeddings.
The Transformer Architecture — Self-Attention, Multi-Head Attention, Positional Encodings, and the Encoder-Decoder structure.
PyTorch / TensorFlow — Become proficient in at least one major deep learning framework.
Recommended Resources
Nerchuko YouTube Channel
Excellent explanations of complex topics in Telugu/English.
Andrej Karpathy: Neural Networks YouTube
Builds a GPT from scratch, explaining every line of code.
2
Working with Large Language Models (LLMs)
8–12 WeeksLearn the practical skills of using and adapting large language models for specific tasks.
Key Topics to Cover
Prompt Engineering — Zero-shot, few-shot, and chain-of-thought prompting to get the best results from LLMs.
Fine-tuning Strategies — Parameter-efficient fine-tuning (PEFT) methods like LoRA and QLoRA to adapt models on custom data.
Retrieval-Augmented Generation (RAG) — Systems that combine LLMs with external knowledge bases using Vector Databases (Chroma, Pinecone).
LLM Application Frameworks — LangChain or LlamaIndex to build complex applications and agentic workflows.
Recommended Resources
LangChain & LlamaIndex Docs Documentation
Essential frameworks for building LLM applications.
OpenAI Cookbook GitHub Repo
Practical examples for using the OpenAI API effectively.
3
Generative Vision & MLOps
6–8 WeeksExpand beyond text to image generation and learn how to deploy your models.
Key Topics to Cover
Diffusion Models — The theory behind models like DALL-E and Stable Diffusion for text-to-image generation.
Multimodal Models — Models like CLIP that can understand both text and images.
Containerization — Package your model and its dependencies into a Docker container for consistent deployment.
API Deployment — Deploy your containerized model as a scalable API endpoint using FastAPI and a cloud service.
Recommended Resources
Krish Naik: MLOps YouTube Series
Complete MLOps playlist covering all stages.
Hugging Face Diffusion Models Course Free Course
A practical course on using and training diffusion models.