Engineers need to communicate effectively when building AI Systems. These terms will help you use a shared vocabulary. This is useful when discussing concepts, reading papers, or collaborating with teammates. Listed in order. 00:00 Agenda 00:28 1. Large Language Model 01:28 2. Tokenization 02:53 3. Vectorization 04:15 4. Attention... 07:22 5. Self-Supervised Learning 12:07 6. Transformer 14:32 7. Fine-tuning 17:05 8. Few-shot Prompting 18:11 9. Retrieval Augmented Generation 20:33 10. Vector Database 23:03 11. Model Context Protocol 25:43 12. Context Engineering 28:17 13. Agents 29:19 14. Reinforcement Learning 34:42 15. Chain of Thought 35:55 16. Reasoning Models 36:36 17. Multi-modal Models 38:21 18. Small Language Models 40:24 19. Distillation 41:47 20. Quantization If you are a software engineer looking to transition to AI, click the link below. AI Engineering Course: #AI #SoftwareEngineering #Agentsshow more

LLM : an AI tool which Generates answers based on the inputs given to it. Tokenization: It's the way LLMs take human sentences as input for their understanding and processing. Vectorization : It's the way LLMs convert the tokens ( words) into numerical embeddings and Store in an array to use it further. Attention : It's the way LLMs relate the Inputs or tokens to give the best answer. Self Supervised Learning: While Attenation LLMs often learn from their experience and predict the answer this process is self Supervised Learning. Transformer = a special deep neural network architecture where inputs pass through many layers of attention + feedforward networks enabling LLMs to deeply understand context and generate answers. Fine Tuning: It's a method of re- training trained models on a particular dataset or knowledge base to make best out of it. Eg a LLM for Medical Practitioners ( it would have fine tuned to answer related to medicine and stuff) Few-Shot Promoting : It's a way by which you can guide LLMs to answer in a specific way by giving examples RAG : RAG is (LLM+ Knowledge+ Vector DB) for better and personalized answers. Vector Database: Similar to Vectorization in Vector Database we store the embeddings which helps LLMs to search and give personalised answers. MCP : Chad Stuff helps LLMs to interact with other external dbs or servers to make our work easy. ( Ask can I buy a PS5 LLMs connect your bank acc through MCP and check PS5 price and tells you to work hard ) Context Engineering : It's the technique by which we can select carefully what all data we are sending into our LLMs saves LLMs from getting unnecessary infos. Agents : Agents are your AI buddy who does a specific task for you. Eg you want to plan a trip you can do it using an AI Agent. Reinforcement Learning : It's the way of feedback driven learning where LLMs decide answers based on users ratings. If a user is liking a particular answer it always try to give similar type and if user downvote it will never give such way of answer. Chain of Thought : It's a prompting technique where we guide LLMs to reason step by step instead of jumping directly to the final answer reduces the chances of errors or hallucinations, especially in tasks that require logic math or multi-step reasoning. Reasoning Models : Unlike general LLMs These are trained for tasks which require high thinking for complex problems. Small Language Models : LLms trained in small data for organisational Policy or requirement eg LLM trained for HR policy. Distillation : LLM with large Datasets teaching LLM with small datasets. It decreases the training times and make work easy. It's cheap ig. Quantization: Compression of LLMs by lowering their precision used in testing process. Do like and give feedback drop a folllow XD.

Thank you sir for clear explanation of the topics …

Great lineup Mastering these terms is like learning the grammar of AI engineering

for more Read

Love this Gaurav... Great stuff 👍 PS: If you could tell me where you sourced this digital board from that would be really helpful. :-)

It's from CorneaWorld. Very good product but the delivery timelines could be better.

Amazing video gaurav! Detailed yet easy to understand.

Very useful video!!

Crazy shit bro, literally on the point no bs . Thnq

Absolutely awesome learning class Gaurav. I have bookmarked your session and will share with my team. The best session I have come across in understanding all AI terms.

G in Gaurav Sen stands for GOAT

Loved it !

Thanks a lot. I knew nothing. Now I know something👍

Nice

Can please explain Hallucination?

Can someone just reply terms with their timestamps and definitions .

This was really great.

Great work....you made it easy

Really helpful