Building Context-Aware Reasoning Applications with LangChain and LangSmith: Harrison Chase

3.5K views · Nov 01, 2023 · 18:53 min · Watch on YouTube ↗

Takeaway

Context delivery method and reasoning pattern (chain vs router vs agent loop) are the two design axes you choose deliberately when building LLM apps.

Summary

Chase (LangChain CEO) breaks 'context' into four delivery channels: instruction prompting (employee handbook), few-shot examples (good for tone and structured output), retrieval-augmented generation (open-book test), and fine-tuning (scaling few-shot to 10k examples).
Reasoning taxonomy across axes: plain code → single LLM call → chain → router (LLM picks branch/tool) → cyclic loop (agent that decides when to stop) → fully autonomous agent (AutoGPT-style).
Argues serious AI products require an engineering layer on top of model APIs because LLMs alone lack current events, code execution, and memory.
Pitches LangChain plus LangSmith as the toolkit to bridge that gap with observability, evals, and tracing across these patterns.

langchainagentsrag

Original description

How can companies best build useful and differentiated applications on top of language models? Many of the products and companies built do this by providing the relevant context to LLMs and asking it to reason appropriately. In this talk, Harrison will discuss the different types of context you should be aware of, the different levels of cognitive architectures that are emerging, and how LangChain and LangSmith are built to help with this journey.

Recorded live in San Francisco at the AI Engineer Summit 2023. See the full schedule of talks at https://ai.engineer/summit/schedule & join us at the AI Engineer World's Fair in 2024! Get your tickets today at https://ai.engineer/worlds-fair

Harrison Chase is the CEO and co-founder of LangChain, a company formed around the open source Python/Typescript packages that aim to make it easy to develop Language Model applications. Prior to starting LangChain, he led the ML team at Robust Intelligence (an MLOps company focused on testing and validation of machine learning models), led the entity linking team at Kensho (a fintech startup), and studied stats and CS at Harvard.