Chroma | Context Engineering

Chroma CEO Jeff Huber sits down with Lance Martin to discuss the current state of agents and more.Find Lance on X, https://x.com/RLanceMartin, and his website, https://rlancemartin.github.io/0:00 Introduction & Welcome0:09 Context Engineering: What It Is and Why It Matters2:05 Context Rot and Performance Degradation3:31 Year in Review: 2025 AI Trends4:28 Giving Agents a Computer (File System & Shell)5:00 Model Context Protocol (MCP) and Tool Bloat6:07 Multi-Tier Action Space Architecture8:24 Tool Search and Progressive Disclosure10:33 Agent Harness Structure & Deep Agents12:17 Skills and Standard Operating Procedures (SOPs)14:13 Context Offloading Techniques15:49 Plan Offloading & The Ralph Wiggum Loop18:00 Context Caching for Cost & Speed18:27 Sub-agents and Context Isolation21:16 Summary: Key Context Engineering Principles22:00 Evolving Context & Continual Learning25:02 Claude Diary: Reflecting on Sessions26:06 Skill Learning from Agent Trajectories27:00 Memory Management in Token Space vs Weights28:35 RLMs: Reason Language Models & Learned Context Management31:30 What Can Be Absorbed Into Models (The Classifier Test)35:30 Memory: Writing vs Retrieval Challenges40:00 File Systems as Agent Primitives42:46 Limitations of File Systems for Large Codebases45:02 Multi-Agent Collaboration & Concurrency Challenges49:35 Layers of Context: Session, Agent, Organizational, Global52:27 File Systems vs Databases: A Hot Take55:51 Sandboxing and Agent Infrastructure58:42 What's Most Exciting: Memory, Personal Agents & Bioscience1:02:18 Wrap Up—Chroma is the open-source AI application database. Batteries included.Embeddings, vector search, document storage, full-text search, metadata filtering, and multi-modal. All in one place.Retrieval that just works. As it should be.Try it today:https://trychroma.com/cloud

Jeff Huber sits down with Drew Breunig to talk about the origins of Context Engineering and more.Drew has a wide range of his writing about AI on his website: https://www.dbreunig.com/00:36 Why write about AI? (Writing as a searchable index)01:28 The two buckets of AI writing: Hype vs. Research04:08 The Gemini 1.5 Paper & Pokemon: The birth of Context Engineering 06:50 The "Karpathy Effect" on Context Engineering08:17 Benchmarks, Model Cards, and the "Agent Harness"11:41 The Weightlifting Metaphor for AI Benchmarks14:02 Testing Opus 4.5: Building internal tools in one shot15:54 Models are untapped: The gains are in the harness17:05 Why isn't there a standard Context Engineering harness?19:20 The "Hello World" Experiment: Testing Agent Frameworks (LangChain, Crew, etc.)21:42 Compact and Grep vs. File Systems23:45 "Naked" tool calls vs. Frameworks24:50 The GPT-8 Thought Experiment: Why Software Engineering still matters27:12 Compound AI Systems: What agents can learn from Data Pipelines31:00 Reliability is the bottleneck (The MAP Report)36:00 Token Speed: When code generates faster than humans can read (Groq/Cerebras)41:00 The UX of Multi-Agent Systems (The "Starcraft" problem)43:57 ChatGPT Deep Research: The Shopping Use Case46:25 Building Trust: Agent Design as Client Services49:30 Continual Learning: Weights vs. Context/Memory51:50 The problem with "Black Box" memory (The Chocolate Example)56:30 The need for "Modes" (Work vs. Home context)--Chroma is the open-source AI application database. Batteries included.Embeddings, vector search, document storage, full-text search, metadata filtering, and multi-modal. All in one place. Retrieval that just works. As it should be.Try it today:https://trychroma.com/cloud

Chroma CEO Jeff Huber chats with Dex Horthy about agents and context engineering.Dex on X: https://x.com/dexhorthy0:00 - Introduction0:23 - Context Engineering Origins0:33 - 12 Factor Agents0:43 - New AI Models1:43 - Model Switching Strategy3:07 - Personal Productivity Systems7:57 - AI UX Patterns13:07 - Todo List Management15:01 - Collaborative AI Workspaces22:09 - In-Person vs Remote24:05 - Tab Complete Patterns25:00 - Shared Context Layer27:00 - Markdown & Airtable32:35 - Data Storage Systems34:00 - AI-Native Organizations36:00 - OAuth & Authentication38:00 - Desktop App Development43:42 - Context Engineering Evolution45:00 - Evals & Observability46:00 - LM-as-Judge Discussion48:00 - Snapshot-Based Evals54:47 - Agent Memory Systems56:00 - Instruction Following Limits57:57 - Closing Remarks--Chroma is the open-source AI application database. Batteries included.Embeddings, vector search, document storage, full-text search, metadata filtering, and multi-modal. All in one place. Retrieval that just works. As it should be.Try it today:https://trychroma.com/cloud