Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Deep dive into LCEL's pipe operator, RunnablePassthrough, RunnableParallel, branching, and fallback patterns for building flexible, declarative AI pipelines in LangChain.
Master the core building blocks of LangChain including chains, prompt templates, language model wrappers, and the LangChain Expression Language for composing AI applications.
Explore LangChain's memory types for building conversational AI — from simple buffer memory to summarization and vector-store-backed long-term memory with persistence strategies.
Learn how to extract structured data from LLM responses using LangChain output parsers — Pydantic models, JSON parsing, format instructions, and retry parsers for robust extraction.
Build end-to-end Retrieval Augmented Generation pipelines with LangChain — covering document loaders, text splitting strategies, vector stores, retrievers, and RAG chain composition.
Langchain_core.tools tool decorator: master LangChain tool creation patterns including the @tool decorator, StructuredTool class, Pydantic input schemas, async tools, and error handling for production-grade agent tools.
A deep comparison of LangChain and the OpenAI Agents SDK covering design philosophy, learning curve, feature depth, and when to choose each framework for production agentic AI systems.
Implement persistence and time travel in LangGraph using MemorySaver, SqliteSaver, and PostgresSaver to checkpoint agent state, replay past executions, and recover from failures.
Build intelligent decision points in LangGraph using conditional edges, router functions, and multi-path branching to create agents that dynamically choose their execution path.
