AI discussions¶

The goal of this section is to get a set of content to support deeper discussions around Gen AI, during chit-chat or interviews.

1. 𝗘𝘅𝗽𝗹𝗮𝗶𝗻 𝗟𝗟𝗠 𝗳𝘂𝗻𝗱𝗮𝗺𝗲𝗻𝘁𝗮𝗹𝘀¶

Cover the high-level workings of models like GPT-3, including transformers, pre-training, fine-tuning, etc.

• General LLM introduction
• Transformer and GPT-3 summary
• How LLM pre-training is done
• How to fine tune existing model
• How RAG works

𝟮. 𝗗𝗶𝘀𝗰𝘂𝘀𝘀 𝗽𝗿𝗼𝗺𝗽𝘁 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴¶

Talk through techniques like demonstrations, examples, and plain language prompts to optimize model performance.

• Prompt Engineering
• Demonstrate prompt engineering
• How to optimize model response performance with prompt

𝟯. 𝗦𝗵𝗮𝗿𝗲 𝗟𝗟𝗠 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗲𝘅𝗮𝗺𝗽𝗹𝗲𝘀¶

Walk through hands-on experiences leveraging models like GPT-3, Langchain, or Vector Databases.

• Review RAG positioning, architecture
• Streamlit app to demonstrate RAG with Chromadb. Offline tool to create vector store and indexing build_agent_domain_rag.py using a Lilian Weng's multi-agents blog.
• Multiple queries RAG with LangChain

𝟰. 𝗦𝘁𝗮𝘆 𝘂𝗽𝗱𝗮𝘁𝗲𝗱 𝗼𝗻 𝗿𝗲𝘀𝗲𝗮𝗿𝗰𝗵¶

Mention latest papers and innovations in few-shot learning, prompt tuning, chain of thought prompting, etc.

• few-shot learning
• chain of thought
• Agentic

𝟱. 𝗗𝗶𝘃𝗲 𝗶𝗻𝘁𝗼 𝗺𝗼𝗱𝗲𝗹 𝗮𝗿𝗰𝗵𝗶𝘁𝗲𝗰𝘁𝘂𝗿𝗲𝘀¶

Compare transformer networks like GPT-3 vs Codex. Explain self-attention, encodings, model depth, etc.

𝟲. 𝗗𝗶𝘀𝗰𝘂𝘀𝘀 𝗳𝗶𝗻𝗲-𝘁𝘂𝗻𝗶𝗻𝗴 𝘁𝗲𝗰𝗵𝗻𝗶𝗾𝘂𝗲𝘀¶

Explain supervised fine-tuning, parameter efficient fine tuning, few-shot learning, and other methods to specialize pre-trained models for specific tasks.

• supervised fine-tuning

𝟳. 𝗗𝗲𝗺𝗼𝗻𝘀𝘁𝗿𝗮𝘁𝗲 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗲𝗻𝗴𝗶𝗻𝗲𝗲𝗿𝗶𝗻𝗴 𝗲𝘅𝗽𝗲𝗿𝘁𝗶𝘀𝗲¶

• From tokenization to embeddings to deployment, showcase your ability to operationalize models at scale, and monitoring model inference.

𝟴. 𝗔𝘀𝗸 𝘁𝗵𝗼𝘂𝗴𝗵𝘁𝗳𝘂𝗹 𝗾𝘂𝗲𝘀𝘁𝗶𝗼𝗻𝘀¶

Inquire about model safety, bias, transparency, generalization, etc. to show strategic thinking.