An interactive course on agent memory.

What agent memory is, why it matters, and why most systems fail.

1. 1.15 minAgent Memory Cheat SheetQuick reference for agent memory primitives, pipeline stages, retrieval modes, schema fields, and deployment configs.
2. 1.19 minWhat is Agent Memory?Agent memory is structured, durable state that lets an LLM agent remember across turns, sessions, and tasks. A primer on episodic, semantic, and procedural memory for engineers.
3. 1.210 minBest Agent Memory Frameworks 2026Survey of agent memory frameworks and what makes them production-ready.
4. 1.218 minWhy Your Agent Forgets (and How to Fix It)Agents forget because most memory systems store everything indiscriminately, then drown the relevant signal. Here's the architecture of memory loss — and what to do about it.
5. 1.319 minThe Cost of Junk MemoriesIndiscriminate memory storage burns tokens, slows retrieval, and degrades answer quality. The math of why filtering before storage beats cleanup after.
6. 1.49 minTyped Memory: Beyond Flat TextFlat text memories collapse facts, preferences, events, entities, and relations into one bucket — and lose temporal and relational signal. The case for typed schemas.
7. 1.510 minMemory vs RAG vs Long ContextThree approaches to giving LLMs more knowledge. They are not interchangeable — each wins on a different axis. A decision framework.

How memories get filtered, extracted, classified, and stored.

1. 2.122 minThe 7-Stage Write PipelinePre-filter, extract, classify, resolve, dedupe, conflict-check, persist. The choreography that turns conversational turns into high-confidence memories.
2. 2.218 minPre-Filter: Rejection Before StorageMost conversational turns are junk. The cheapest filter — regex and length heuristics — rejects them before any LLM extraction runs.
3. 2.318 minLLM Extraction as Filtering, Not BrainExtraction prompts that ask 'what is memorable here' beat prompts that ask 'extract every fact'. The framing flips the precision/recall trade-off.
4. 2.421 minEntity Resolution: From Pronouns to IdentityHe, she, my boss, that thing we discussed — agents must turn references into stable identities. A four-stage cascade from grammar to LLM judge.
5. 2.520 minThree Tiers of DeduplicationHash-equality, cosine-similarity, LLM-judge. Three thresholds that catch 99% of duplicates while paying LLM costs only when needed.

How memories get retrieved, fused, ranked, and assembled into context.

1. 3.111 minFive Retrievers Are Better Than OneSemantic, lexical, entity-graph, temporal, type-filter. No single retriever wins all queries — fusion does.
2. 3.218 minReciprocal Rank Fusion, ExplainedRRF combines ranked lists without needing to normalize scores. The formula, the intuition, and why it beats weighted score sums.
3. 3.310 minThe Query Optimizer for MemoryNot every query needs every retriever. Routing by entity density, temporal precision, and lexical rarity halves p99 latency.
4. 3.410 minEntity Graphs for Multi-Hop ReasoningVectors find similar memories. Graphs find connected ones. Why agent memory needs both — and how to score multi-hop paths.
5. 3.510 minContext Aggregation: Token BudgetingSix categories, one budget. Allocating tokens between facts, preferences, events, entities, summaries, and recent turns — and avoiding lost-in-the-middle.

Confidence, decay, fusion — the formulas behind quality memory.

1. 4.114 minThe Confidence FormulaSource strength, repetition, extractor quality, type prior. A weighted blend that turns 'how much do we trust this memory' into a single number.
2. 4.214 minWhy Logarithmic Repetition BoostThe 100th observation should not outweigh the 10th. Why log scaling beats linear for repetition signals — with the curve.
3. 4.314 minFreshness Decay CurvesFacts age slowly. Preferences age faster. Events expire. Type-specific exponential decay with retrieval-driven half-life resets.
4. 4.414 minBM25 for Memory, in Plain EnglishTerm frequency saturation, inverse document frequency, length normalization. Why BM25 still belongs in your retrieval mix in 2026.
5. 4.510 minHNSW Tuning: m, ef, and Memory CostThree parameters control the speed/recall/memory triangle for vector indexes. The math you need to size an HNSW index correctly.

Hallucination defense, drift detection, scale, maintenance.

1. 5.110 minThree Layers of Hallucination DefenseWrite-time grounding, store-time consistency, read-time faithfulness. Defense in depth for memories that must not lie.
2. 5.211 minDetecting Memory DriftConcept drift, data drift, schema drift, vocabulary shift. The four kinds of drift in long-running memory — and how to detect them.
3. 5.39 minConcept Drift: Dual-Signal DetectionCentroid distance alone is noisy. Relation overlap alone is sparse. Together, they reliably separate entity evolution from semantic shift.
4. 5.410 minScaling to 1B Memories: Index TiersOne shard at 100K. Multiple thread-per-core shards on one node to 100M. A multi-node sharded cluster beyond. Brain's scaling ladder and where each rung breaks.
5. 5.58 minThe Background Worker: 7 Maintenance JobsDecay, consolidation, drift scan, consistency check, embedding refresh, garbage collect, snapshot. The maintenance loop that keeps memory healthy.
6. 5.68 minTop Mistakes Building Agent MemoryCommon pitfalls and lessons learned while building agent memory systems into production.
7. 5.710 minSchema Design Patterns for Agent MemoryTyped memory schemas, entity relationships, and field design for production agent systems.
8. 5.89 minLong-Term Memory for AgentsPersistence strategies, durability guarantees, and retrieval patterns for agent memory that outlives sessions.

How agent memory applies to specific industries and domains.

1. 6.18 minAgent Memory for Customer SupportHow agent memory enables support agents to remember tickets, preferences, and resolution patterns for faster, context-aware responses.
2. 6.28 minAgent Memory for Research AutomationHow agent memory enables research agents to remember papers, findings, and hypotheses to avoid redundant analysis and discover patterns across thousands of sources.
3. 6.38 minAgent Memory for Autonomous Trading AgentsHow agent memory enables trading agents to remember market patterns, trade outcomes, and risk events to improve win rates and avoid repeating losses.
4. 6.49 minAgent Memory for Healthcare & Medical AIHow agent memory enables medical agents to remember patient history, diagnoses, and medications for comprehensive, continuity-of-care responses without doctors re-reading past records.
5. 6.59 minAgent Memory for Legal Document AnalysisHow agent memory enables legal AI to remember case facts, precedents, and client decisions for faster brief drafting and consistent legal arguments.
6. 6.68 minAgent Memory for E-commerce PersonalizationHow agent memory enables shopping agents to remember customer purchase history and preferences for personalized recommendations and higher cart conversion.
7. 6.79 minAgent Memory for Financial ServicesHow agent memory enables portfolio and trading agents to remember investment positions, risk rules, and market events for consistent decision-making across months of active trading.
8. 6.89 minAgent Memory for HR & RecruitmentHow agent memory enables recruitment agents to remember candidate profiles, interview feedback, and role requirements for faster hiring decisions and consistent candidate experience.
9. 6.910 minAgent Memory for Content GenerationHow agent memory enables content generation agents to remember brand voice, prior outputs, and audience preferences to maintain consistency and reduce brand voice drift across campaigns.

Using Brain with LangChain, LlamaIndex, AutoGen, Claude, and other frameworks.

1. 7.17 minUsing Brain with LangChainIntegrate Brain as a persistent memory layer into LangChain agents to reduce redundant LLM calls by 40% with learned context.
2. 7.27 minUsing Brain with LlamaIndexIntegrate Brain as a memory layer in LlamaIndex RAG systems to achieve 35% faster subsequent queries via learned context.
3. 7.38 minUsing Brain with AutoGenIntegrate Brain into AutoGen multi-agent systems to enable shared memory and improve multi-agent coordination by 50%.
4. 7.47 minUsing Brain with ClaudeIntegrate Brain to manage context and memory with the Claude API, reducing context window usage by 50% via selective memory.

Deployment checklists, cost optimization, schema decisions, and retrieval patterns.

1. 8.18 minMemory Monitoring & ObservabilityHow to monitor agent memory systems in production using key metrics, alerting, and logging patterns to ensure reliability and performance.
2. 8.26 minProduction Deployment ChecklistPre-deploy, deploy-day, and post-deploy verification items for memory systems to ensure smooth rollouts and rapid rollback capability.
3. 8.38 minCost Optimization for Agent MemoryFive strategies to reduce memory system costs while maintaining performance and data quality.
4. 8.48 minChoosing Your Memory SchemaDecision tree and comparison for selecting the right memory types—fact vs. event vs. entity vs. relation—based on your agent's needs.
5. 8.57 minSelf-Hosted vs Managed CloudDecision guide for choosing the right memory deployment model—self-hosted open-source server (single node or HA cluster) or managed cloud—based on latency, cost, ops overhead, and data sovereignty.
6. 8.68 minRetrieval Strategy SelectorDecision guide for choosing retrieval strategies
7. 8.79 minTemporal Context WindowsTime-based memory management and retrieval patterns
8. 8.89 minMemory Consolidation PatternsStrategies for consolidating, summarizing, and archiving memories

Brain vs alternative memory and retrieval systems.

1. 9.114 minMem0 vs Letta vs Zep vs BrainAn honest comparison of the four leading agent memory systems — what each is good at, where each falls short, and which one fits which use case.
2. 9.210 minVector DB ≠ Memory: Why Pinecone Isn't EnoughA vector database is a substrate. Agent memory is a system on top of one. Why you need types, supersession, and temporal reasoning — not just embeddings.
3. 9.312 minLangChain Memory vs LangGraph State vs BrainConversational buffers, graph state, durable typed memory. Three different abstractions for three different jobs.
4. 9.48 minBrain vs PineconeFeature-by-feature comparison of Brain and Pinecone for vector search and semantic memory.
5. 9.57 minBrain vs PostgreSQL + pgvectorWhen to use Brain agent memory vs pgvector for vector storage in agents.
6. 9.67 minBrain vs Redis VectorComparing Brain agent memory to Redis Vector (formerly RediSearch) for vector workloads.
7. 9.77 minAlternatives to Mem0Comparing Brain to Mem0 and other agent memory frameworks.
8. 9.810 minBrain vs. Vector DatabasesUnderstand when to use an agent memory layer (Brain) vs. a vector database for AI systems.
9. 9.98 minMigrating from Mem0 to BrainAPI mapping and migration steps from Mem0 to Brain
10. 9.108 minMigrating from Pinecone to BrainWhen and how to migrate from vector DB to agent memory
11. 9.115 minOpen Source to Cloud Upgrade GuideZero-code upgrade from open-source Brain to Managed Cloud