详细内容
Build Agents That Recall |
Buildfast,accurate, andpersonalizedagents with the only platform that systematically engineersrelevant contextfrom chat history and business data.
Agent Memory | Graph RAG | Context Assembly
Get Started Free Try Playground
The Foundational Memory Layer for AI from Zep AI on Vimeo
Playing in picture-in-picture
Unmute
Like
Share
Pause
00:00
00:01
CC/subtitlesSettingsTranscript
Quality1080p
SpeedNormal
CC/subtitlesEnglish (United States)
Fullscreen
Zep is a memory layerfor AI Agents.
Trusted by developers at
What is Context Engineering?
Context Engineering assembles relevant information around an LLM for reliable task completion. Unlike basic prompting, it dynamically integrates user preferences, conversation history, and business data.
Agents often fail due to missing personalized context. Zep solves this by automatically creating temporal knowledge graphs to organize and retrieve context for each interaction.
Learn about Context Engineering
APPLICATION & INTERFACE
AGENT FRAMEWORK
CONTEXT ENGINEERING
ZEP CAPABILITIES
Context Orchestration
Agent Memory
Graph RAG
Search Tools
Context Assembly
Tool Integration
LLM
tobi lutke@tobi·Jun 18, 2025
I really like the term "context engineering" over prompt engineering. It describes the core skill better: the art of providing all the context for the task to be plausibly solvable by the LLM.
34121008,300
Andrej Karpathy@karpathy·Jun 25, 2025
+1 for "context engineering" over "prompt engineering". People associate prompts with short task descriptions you'd give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window
522180013,600
How Zep Works
Zep transforms conversations, business data, & user interactions into a living knowledge graph that evolves with every interaction. Relevant and accurate context is automatically assembled into a context block when you need it.
Unified, dynamic, long-term context store for agents (Zep's knowledge graph) - YouTube
Zep AI
872 subscribers
Unified, dynamic, long-term context store for agents (Zep's knowledge graph)
Zep AI
Search
Watch later
Share
Copy link
Info
Shopping
Tap to unmute
If playback doesn't begin shortly, try restarting your device.
More videos
More videos
You're signed out
Videos you watch may be added to the TV's watch history and influence TV recommendations. To avoid this, cancel and sign in to YouTube on your computer.
CancelConfirm
Share
Include playlist
An error occurred while retrieving sharing information. Please try again later.
0:00
0:00 / 2:47
•Live
•
1
Graph Construction
Automatic extraction of entities, relationships, and facts from conversations and business data, reconciling new information with existing data to maintain accuracy.
2
Relevant Retrieval
When your agent needs context, Zep searches the knowledge graph and returns the most relevant information.
3
Context Assembly
Zep delivers structured, LLM-ready context combining user traits, interactions, and business data while being token efficient.
Product Features
R
Robbie
2024-09-07 14:27
I only wear Adidas shoes. I love them!
Facts
- Robbie only wears Adidas shoes.
- Robbie strongly favors Adidas shoes.
R
Robbie
2024-10-14 09:12
My shoes fell apart and I need to return them. I'm super angry! I'll be wearing Nike going forward!
Facts
- Robbieonly wears Adidasshoes.
- Robbiestrongly favors Adidasshoes.
- Robbie'sAdidas shoes fell apart.
- Robbie needs to return their shoes.
- Robbie is angryabout their Adidas shoes falling apart.
- Robbie intends to wearNikeshoes in the future.
Agent Memory
Deploy Agents with Perfect Memory
Your agents remember user preferences, past conversations, and business context across all interactions. No more starting from scratch or losing important details.
Graph RAG
Super Fast Graph RAG for Dynamic Data
Connect agents to your business data with Graph RAG that understands relationships and context, and automatically handles dynamic data. Your agents access the right information at the right time, in milliseconds.
Copy
FACTS and ENTITIES provide relevant context for the current conversation.
# Key facts with their date ranges
# format: FACT (Date range: from - to)
<FACTS>
- Emily Painter cannot log in. (2024-11-14 02:13:19+00:00 - present)
- Account Emily0e62 is suspended due to payment failure. (2024-11-14 02:03:58+00:00 - present)
- Payment failed using card ending in 1234. (2024-09-15 00:00:00+00:00 - present)
- Failure reason: Card expired. (2024-09-15 00:00:00+00:00 - present)
</FACTS>
# Key entities and their descriptions
# ENTITY_NAME: entity summary
<ENTITIES>
- Emily0e62: User account owned by Emily Painter, currently suspended due to payment failure and login issues.
- Card expired: Reason for the failed payment transaction.
</ENTITIES>
Context Assembly
Automated Context Creation
Stop manually crafting prompts. Zep automatically assembles relevant user memory and business context into optimized prompts for reliable agent performance.
Trusted by Industry Leaders
See how teams are transforming their AI applications with Zep
“Zep is one of the most exciting things I've seen for real-world agent use cases in a long time. Their innovative approach is truly game-changing.”
Ken Collins
VP of Product, Torq and GenAI Expert
“Zep just introduced a game-changing way for AI agents to remember and learn. Unlike other systems that only retrieve static documents, Zep uses a temporal knowledge graph to combine conversations and structured business data, keeping track of how things change over time.”
Lior Sinclair
Founder/CEO, AlphaSignal
“Zep AI was instrumental in enabling the Sidekick's personalized experience through dynamic memory retrieval. Their innovative tech stack is powering groundbreaking projects like ArtPrize 2024, taking personalized AI experiences to the next level.”
Mark Losey
CTO at Flockx
“Zep AI empowers AI systems to think and remember like humans. By organizing memories into structured episodes and extracting key insights, it builds smarter, more intuitive AI agents that revolutionize how businesses harness intelligence.”
Vijay Morampudi
Senior Director - AI CoE, Axtria
“Zep is one of the most exciting things I've seen for real-world agent use cases in a long time. Their innovative approach is truly game-changing.”
Ken Collins
VP of Product, Torq and GenAI Expert
“Zep just introduced a game-changing way for AI agents to remember and learn. Unlike other systems that only retrieve static documents, Zep uses a temporal knowledge graph to combine conversations and structured business data, keeping track of how things change over time.”
Lior Sinclair
Founder/CEO, AlphaSignal
“Zep AI was instrumental in enabling the Sidekick's personalized experience through dynamic memory retrieval. Their innovative tech stack is powering groundbreaking projects like ArtPrize 2024, taking personalized AI experiences to the next level.”
Mark Losey
CTO at Flockx
“Zep AI empowers AI systems to think and remember like humans. By organizing memories into structured episodes and extracting key insights, it builds smarter, more intuitive AI agents that revolutionize how businesses harness intelligence.”
Vijay Morampudi
Senior Director - AI CoE, Axtria
Built for Teams, Proven at Scale
Deploy personalized agents in days, not months. Enterprise-grade compliance meets developer-friendly APIs.
For Developers
Skip building complex infrastructure. Three lines of code to production.
Copy
# Add conversation to memory
zep.memory.add(thread_id, messages)
# Get relevant context
memory = zep.memory.get(thread_id)
context = memory.context
User memory with three lines of code
Ingest business data as JSON, text, or messages
Works with LangChain, LangGraph, and popular frameworks
For Engineering Leaders
Measurable performance improvements with enterprise compliance.
Enterprise compliance with SOC2 Type 2 and HIPAA
Days vs. months implementation without hiring scarce AI talent
100%+ accuracy improvements through personalized context
For Engineering Leaders Schedule Demo
Customized for Your Domain
Zep adapts to your business through custom entity types and relationship models. These models enable precision recall of exactly the required context, so your agents understand your business domain, not just generic conversations.
Sales and MarketingCustomer SupportE-commerceEducationHealthcare
Sales and Marketing
Store lead preferences, product interests, campaign interactions, and buying signals.
Your sales agents understand prospect history, pricing discussions, and engagement patterns to personalize outreach and close deals faster.
sales_and_marketing_entities.py
Copy
class Lead(EntityModel):
"""Represents a sales lead or prospect."""
company_size = Field(
description="startup, SMB, mid-market, enterprise"
)
budget_range = Field(
description="Budget discussed or indicated"
)
decision_timeline = Field(
description="Expected decision timeframe"
)
Zep is the State of the Art in Agent Memory
Zep is the current state-of-the-art in agent memory, excelling in the LongMemEval benchmark, a challenging evaluation that closely models enterprise use cases.
100%+
Accuracy Improvements
Agents perform better when provided with the right context at the right time.
90%
Latency Reduction
Optimized context retrieval delivers the right information without overwhelming LLMs with irrelevant data.
98%
Token Efficiency
Smart context assembly reduces token usage while maintaining comprehensive understanding.
Graphiti Graph Framework
Powered by Open-Source
Zep's open-source temporal knowledge graph library, Graphiti, serves as the core of our capability to swiftly integrate new data streams and provide comprehensive historical context regarding user states.
Ready to Build Smarter Agents?
Join developers and engineering teams using Zep to deploy agents that understand their users and business context.
Get Started Free Talk to Sales
Start with our free tier
No credit card required • Full API access • Community support
\
\
\
\