Developer docs
API playgroundTry for free, no card

Search company profiles

GibsonAI

Full company profile

uuid0000wjc

Namestring
GibsonAI
Legal namestring
Memori Labs Inc.
Websiteurl
gibsonai.com
Company typeenum
Private
Founded yearint
2024
Descriptiontext

GibsonAI, operating under the DBA "Memori" as the legal entity Memori Labs Inc., is a New York-based private company founded in 2024 that builds agent-native memory infrastructure for production AI systems. The core product, Memori, is a LLM-agnostic layer that captures each agent chat turn and automatically classifies it into facts, preferences, rules, and summaries, then performs targeted recall and selective semantic search to inject only relevant context on demand. Supporting modules include Memory Graph (entity relationship visualization), Memori Analytics (usage and cache hit rate monitoring), Memori Cloud (hosted deployment launched 2025), and Payments Vault (PCI/SOC 2-compliant handling of payment data and PII). Published benchmark results report 81.95% accuracy on the LoCoMo benchmark with a 95% reduction in token usage versus full-context retrieval.

The company serves two primary segments: AI agent developers and builders integrating memory into production systems, and enterprise AI deployers requiring secure, scalable memory infrastructure with cost controls. A secondary segment targets robotics and IoT use cases. The go-to-market combines product-led growth (one-line SDK integration, free tier, sub-minute onboarding) with community-led motion (open-source GitHub SDK, Discord, blog, benchmark research) and direct enterprise sales for larger deployments. Revenue is generated through a freemium model with subscription-based Pro tier, usage-based scaling on paid plans, and a startup program offering 3-month free Pro access. Notable disclosed partnerships and endorsements include MongoDB, Cockroach Labs, DigitalOcean, SK Networks (NAMUHx wellness robots), Swarms Corporation, and Agno.

The company is in an early commercial stage with 1-10 employees and has raised a single disclosed funding round — a $3.5M seed in May 2024 led by Oceans with participation from f7 Ventures, RiverPark Ventures, and Struck Capital. No revenue, ARR, or paying customer count is publicly disclosed. Compliance posture includes PCI DSS and SOC 2 Type 2 certifications, positioning the company for regulated enterprise workloads ahead of typical early-stage timelines.

Short descriptiontext

GibsonAI, operating as Memori Labs Inc. (DBA Memori), builds LLM-agnostic memory infrastructure that captures, classifies, and retrieves structured context for production AI agents, reducing LLM token costs by 95%. It serves AI agent developers and enterprise AI deployers via a freemium SDK, Memori Cloud, and direct sales.

Operating statusenum
Operating
Ownership categoryenum
Headcount rangeband
1–10
akta.pro rankint
HeadquartersSan Francisco, United States
HQ citystring
San Francisco
HQ countrystring
United States
HQ regionstring
North America
Markets served

Serves global market

Keyword5 values
AI memory infrastructure, agent memory layer, LLM context management, AI agent platform, semantic memory search
Industry3 codes
PRIMARYHDAEANAD
LLMOps & Generative AI Platforms (Prompt/Agent Orchestration, RAG)
Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise
SECONDARYHDAAAAAF
AI Memory & Storage for Training (HBM/DDR/SSD/Object Storage)
Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/AI Infrastructure & Compute (GPUs/ASICs, accelerators)
SECONDARYHDAEAAAN
Database-as-a-Service (DBaaS) Platforms
Hardware, Technology & Telecom/Enterprise Platforms & Databases/Database Management Systems (Relational & NoSQL)
NAICS code1 code
  • Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services518
SIC code1 code
  • Services-Prepackaged Software7372
Product category
AI Memory Infrastructure
GTM motion2 records

Each record includes

Type, Description, Source

Revenue model2 records
1Memori Cloud Subscription
TypeSubscription Recurring
Description

Cloud-hosted memory storage and search service with tiered plans including Pro tier for startups (3-month free access program available). Usage-based scaling for production workloads.

memorilabs.ai
2Self-serve SDK
TypeUsage Based
Description

Developer SDK with free signup tier, enabling tokenless recall and structured memory. Revenue generated through usage-based consumption on paid tiers.

memorilabs.ai
Marketing channels8 records

Each record includes

Title, Type, Stage, Description, Source

Distribution channels4 records

Each record includes

Title, Type, Scope, Target buyer, Description, Source

Cost components5 values
Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations
Pricing details2 tiers
1Free tier with basic memory infrastructure
ModelFreemiumBilling cadenceMonthly
Notes

Free signup includes basic SDK access and starter credits for testing

memorilabs.ai
2Pro plan for growing teams
ModelSubscriptionBilling cadenceMonthly
Notes

3-month free access available through startup program

memorilabs.ai
GTM typeB2B
B2B
Offering typeSoftware
Software
Core offering1 text field

GibsonAI (operating as Memori Labs Inc., DBA Memori) provides an LLM-agnostic memory infrastructure layer that turns AI agent execution and conversation into structured, persistent state for production systems. Its product captures each chat turn, classifies content into facts, preferences, rules, and summaries, and serves targeted recall with semantic search to reduce LLM token costs. The offering includes a Python SDK for drop-in integration and a hosted Memori Cloud service.

Differentiator
Functional benefit
Problem solved
Quantifiable outcome1 of 3 values shown
  • 81.95% accuracy on LoCoMo benchmark while reducing token usage by 95%
+2 more records
Product overview1 text field

Memori Labs offers a unified platform for agent-native memory infrastructure. The core product is Memori, a LLM-agnostic memory layer that transforms agent conversations into structured, persistent state. The platform includes Memory Graph for visualizing entity relationships, Memori Analytics for monitoring performance metrics, and Payments Vault for secure handling of payment data and PII. Memori Cloud provides a hosted deployment option with managed infrastructure, semantic search, and graph relationship capabilities. Organizations can self-host by storing memory in their own database while layering on Memori's hosted capabilities.

Product and service2 records
1Memori
CategoryAI Memory Infrastructure SDK
Description

LLM-agnostic memory infrastructure SDK that captures AI agent chat turns, classifies them into facts, preferences, rules, and summaries, and provides targeted recall and semantic search across conversations and documents. Designed for developers building AI agents and enterprises deploying production AI systems requiring persistent memory.

2Memori Cloud
CategoryCloud-Hosted AI Memory Service
Description

Hosted cloud deployment of the Memori memory layer providing instant memory storage and search with managed infrastructure, semantic search, memory graph relationships, analytics, and third-party integrations. Targets teams that prefer managed infrastructure over self-hosting the SDK.

Scale indicator3 records

Each record includes

Type, Value, Description, Source

Partnership3 partners
Strategic tierCoreTypeTechnology or Integration
Description

MongoDB is exploring integration of Memori's memory abstraction with MongoDB's flexible document model and integrated vector search. Andrew Davidson, SVP Products, cited the combination as a powerful approach for richer, more dynamic context in LLM-driven applications.

Strategic tierCoreTypeTechnology or Integration
Description

CockroachDB integration enables developers to spin up memory-backed AI workflows with minimal cost and friction. Spencer Kimball, Co-founder & CEO, endorsed the integration for testing and production scaling without upfront database costs.

Strategic tierCoreTypeTechnology or Integration
Description

SK Networks uses Memori for NAMUHx wellness robots, providing persistent memory layer for truly personalized wellness experiences. Simon Hong, Head of AI Innovation, cited Memori's strong recall accuracy and inference-cost savings as key factors.

Recent move6 records

Each record includes

Date, Type, Title, Description, Source

Expansion highlight7 records

Each record includes

Type, Description

Peers10 records
TypeDirect peer
Description

Mem0 is a direct competitor offering a memory layer for LLM applications and AI agents, with automatic memory extraction, storage and retrieval across sessions. It targets the same developer persona, uses a similar self-serve/SDK model and competes head-to-head on accuracy benchmarks and token-cost reduction.

2Zep (GetZep)
TypeDirect peer
Description

Zep provides long-term memory and a temporal knowledge graph for AI assistants and agents, with APIs for message storage, summarization and fact extraction. It overlaps closely with Memori on agent memory infrastructure and on the use of structured recall rather than raw context stuffing.

TypeDirect peer
Description

Letta (formerly MemGPT) builds persistent memory and stateful agent infrastructure for LLM applications. It competes directly on long-horizon agent memory, context management and developer-facing APIs for production deployments.

4Cognee
TypeDirect peer
Description

Cognee offers an open-source memory layer for AI applications that structures and retrieves knowledge from conversations and documents using graph and vector storage. It is comparable to Memori on the agent-memory abstraction, structured recall and self-hostable architecture.

TypeBroad incumbent
Description

LangChain is the dominant LLM application framework with its own agent orchestration, memory abstractions and integrations across hundreds of vector stores. It is a broad incumbent whose native memory components overlap with Memori's value proposition for the same developer audience.

TypeBroad incumbent
Description

LlamaIndex provides data frameworks for LLM applications including indexing, retrieval and agent memory components. It is a broad incumbent in the same ecosystem, offering overlapping capabilities around structured context and agent-side recall.

TypeBroad incumbent
Description

Pinecone is a leading managed vector database used for retrieval-augmented generation and agent memory use cases. It is a broad incumbent with significantly more capital and a larger installed base that could expand into the agent-memory category Memori occupies.

TypeBroad incumbent
Description

Weaviate is an open-source vector database with hybrid search, modules and a managed cloud service widely used for RAG and agent memory. It is comparable on the retrieval and semantic-search backbone that Memori layers its agent-memory product on top of.

TypeEmerging player
Description

Chroma is an open-source embedding database popular with AI developers for RAG and lightweight agent memory. It overlaps with Memori on the developer-targeted self-serve distribution and on serving as a retrieval layer underneath agent applications.

TypeEmerging player
Description

Qdrant is an open-source vector similarity search engine with a managed cloud offering, used for semantic search and memory in LLM applications. It is comparable as a retrieval infrastructure layer that Memori's customers and competitors rely on for agent context.

Market position
Strengths5 records

Each record includes

Headline, Details, Source

Weaknesses5 records

Each record includes

Headline, Details, Source

Competitive moat5 records

Each record includes

Type, Details

Key risks6 records

Each record includes

Headline, Details, Source

Key highlights6 records

Each record includes

Headline, Details, Source

Customer concentration

Classification, Details

Named customers6 records

Each record includes

Name, Industry, Type, Use case, Source, UUID

Segment3 records

Each record includes

Title, Type, Primary, Description, Pain point addressed, Use case, Source

Ideal customer profile3 records

Each record includes

Profile, Firmographic size, Sales motion, Sales cycle length, Buying structure, Purchase trigger, Buyer persona, Geography, Industry vertical, Primary use case, Description, Pain points, Evidence proof points, Target buyer

Technology focused
Yes
API detail
Has APIbool
Yes

Docs URL, Description

Integration2 records

Each record includes

Title, Type, Description, Source

AI capability4 records

Each record includes

Type, Description, Source

AI maturity
App detail

Has app

Feature9 records

Each record includes

Title, Differentiator, Description, Source

Core technology
Revenue estimate
Valuation estimate
Number of profiles
No data
Compliance2 records

Each record includes

Name, Class, Description

Funding overview

Funding stage, Last funding date, Total funding USD

Funding rounds1 record

Each record includes

Round, Amount USD, Date, Pre money valuation, Total investors, Investors, News

Investors4 records

Each record includes

Name, Type, Date of entry, Rounds participated, Website

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Each record includes

Name, Acquisition type, Announced date, Completed date, Status, Website, News

Investment

Each record includes

Name, Round, Announced date, Lead investor, Website, News

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

GibsonAI

AI Memory Infrastructuregibsonai.com

GibsonAI, operating as Memori Labs Inc. (DBA Memori), builds LLM-agnostic memory infrastructure that captures, classifies, and retrieves structured context for production AI agents, reducing LLM token costs by 95%. It serves AI agent developers and enterprise AI deployers via a freemium SDK, Memori Cloud, and direct sales.

What GibsonAI does

GibsonAI, operating under the DBA "Memori" as the legal entity Memori Labs Inc., is a New York-based private company founded in 2024 that builds agent-native memory infrastructure for production AI systems. The core product, Memori, is a LLM-agnostic layer that captures each agent chat turn and automatically classifies it into facts, preferences, rules, and summaries, then performs targeted recall and selective semantic search to inject only relevant context on demand. Supporting modules include Memory Graph (entity relationship visualization), Memori Analytics (usage and cache hit rate monitoring), Memori Cloud (hosted deployment launched 2025), and Payments Vault (PCI/SOC 2-compliant handling of payment data and PII). Published benchmark results report 81.95% accuracy on the LoCoMo benchmark with a 95% reduction in token usage versus full-context retrieval.

The company serves two primary segments: AI agent developers and builders integrating memory into production systems, and enterprise AI deployers requiring secure, scalable memory infrastructure with cost controls. A secondary segment targets robotics and IoT use cases. The go-to-market combines product-led growth (one-line SDK integration, free tier, sub-minute onboarding) with community-led motion (open-source GitHub SDK, Discord, blog, benchmark research) and direct enterprise sales for larger deployments. Revenue is generated through a freemium model with subscription-based Pro tier, usage-based scaling on paid plans, and a startup program offering 3-month free Pro access. Notable disclosed partnerships and endorsements include MongoDB, Cockroach Labs, DigitalOcean, SK Networks (NAMUHx wellness robots), Swarms Corporation, and Agno.

The company is in an early commercial stage with 1-10 employees and has raised a single disclosed funding round — a $3.5M seed in May 2024 led by Oceans with participation from f7 Ventures, RiverPark Ventures, and Struck Capital. No revenue, ARR, or paying customer count is publicly disclosed. Compliance posture includes PCI DSS and SOC 2 Type 2 certifications, positioning the company for regulated enterprise workloads ahead of typical early-stage timelines.

GibsonAI firmographics

Firmographics
Name
GibsonAI
Legal name
Memori Labs Inc.
Website
https://gibsonai.com
Company type
Private
Founded year
2024
Operating status
Operating
Headcount range
1–10 employees
Short description
GibsonAI, operating as Memori Labs Inc. (DBA Memori), builds LLM-agnostic memory infrastructure that captures, classifies, and retrieves structured context for production AI agents, reducing LLM token costs by 95%. It serves AI agent developers and enterprise AI deployers via a freemium SDK, Memori Cloud, and direct sales.
Ownership category
akta.pro rank

GibsonAI industry classification

Industry
Product category
AI Memory Infrastructure
NAICS
Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (518)
SIC
Services-Prepackaged Software (7372)
akta.pro industry
PRIMARYHDAEANAD
LLMOps & Generative AI Platforms (Prompt/Agent Orchestration, RAG)
Hardware, Technology & Telecom/Enterprise Platforms & Databases/Artificial Intelligence Platforms for Enterprise
SECONDARYHDAAAAAF
AI Memory & Storage for Training (HBM/DDR/SSD/Object Storage)
Hardware, Technology & Telecom/Artificial Intelligence & Machine Learning/AI Infrastructure & Compute (GPUs/ASICs, accelerators)
SECONDARYHDAEAAAN
Database-as-a-Service (DBaaS) Platforms
Hardware, Technology & Telecom/Enterprise Platforms & Databases/Database Management Systems (Relational & NoSQL)

Keywords

  • AI memory infrastructure
  • Agent memory layer
  • LLM context management
  • AI agent platform
  • Semantic memory search

Where GibsonAI is headquartered

Location

Headquarters

HQ city
San Francisco
HQ country
United States
HQ region
North America

Markets served

GibsonAI business model

Business model
GTM type
B2B
Offering type
Software
Cost components
Technology or R&D, Personnel, Infrastructure, Marketing or Sales, Operations

Revenue model

  1. Memori Cloud Subscription: Cloud-hosted memory storage and search service with tiered plans including Pro tier for startups (3-month free access program available). Usage-based scaling for production workloads.
  2. Self-serve SDK: Developer SDK with free signup tier, enabling tokenless recall and structured memory. Revenue generated through usage-based consumption on paid tiers.

Pricing tiers

ModelBillingPrice
FreemiumMonthlyFree tier with basic memory infrastructure
SubscriptionMonthlyPro plan for growing teams

Go-to-market motion2 records

Distribution channels4 records

Marketing channels8 records

GibsonAI product offering

Product offering

Core offering

GibsonAI (operating as Memori Labs Inc., DBA Memori) provides an LLM-agnostic memory infrastructure layer that turns AI agent execution and conversation into structured, persistent state for production systems. Its product captures each chat turn, classifies content into facts, preferences, rules, and summaries, and serves targeted recall with semantic search to reduce LLM token costs. The offering includes a Python SDK for drop-in integration and a hosted Memori Cloud service.

Product overview

Memori Labs offers a unified platform for agent-native memory infrastructure. The core product is Memori, a LLM-agnostic memory layer that transforms agent conversations into structured, persistent state. The platform includes Memory Graph for visualizing entity relationships, Memori Analytics for monitoring performance metrics, and Payments Vault for secure handling of payment data and PII. Memori Cloud provides a hosted deployment option with managed infrastructure, semantic search, and graph relationship capabilities. Organizations can self-host by storing memory in their own database while layering on Memori's hosted capabilities.

Differentiator

Problem solved

Functional benefit

Products and services

  • Memori LLM-agnostic memory infrastructure SDK that captures AI agent chat turns, classifies them into facts, preferences, rules, and summaries, and provides targeted recall and semantic search across conversations and documents. Designed for developers building AI agents and enterprises deploying production AI systems requiring persistent memory.
  • Memori Cloud Hosted cloud deployment of the Memori memory layer providing instant memory storage and search with managed infrastructure, semantic search, memory graph relationships, analytics, and third-party integrations. Targets teams that prefer managed infrastructure over self-hosting the SDK.

Quantifiable outcome

  • 81.95% accuracy on LoCoMo benchmark while reducing token usage by 95%
  • +2 more outcomes

Companies that use GibsonAI

Customer profile

Named customers6 records

Segments3 records

Ideal customer profiles3 records

GibsonAI technology and API

Technology

Technology focussed Yes

API detail

Has API
Yes
API docs
API detail

Core technology

AI maturity

App detail

Integration2 records

AI capability4 records

Feature9 records

GibsonAI partnerships and signals

Strategic signal

Partnerships

Three partnerships are on record, tiered core.

  • MongoDBcoreTechnology or IntegrationMongoDB is exploring integration of Memori's memory abstraction with MongoDB's flexible document model and integrated vector search. Andrew Davidson, SVP Products, cited the combination as a powerful approach for richer, more dynamic context in LLM-driven applications.
  • Cockroach LabscoreTechnology or IntegrationCockroachDB integration enables developers to spin up memory-backed AI workflows with minimal cost and friction. Spencer Kimball, Co-founder & CEO, endorsed the integration for testing and production scaling without upfront database costs.
  • SK NetworkscoreTechnology or IntegrationSK Networks uses Memori for NAMUHx wellness robots, providing persistent memory layer for truly personalized wellness experiences. Simon Hong, Head of AI Innovation, cited Memori's strong recall accuracy and inference-cost savings as key factors.

Scale indicators3 records

Recent moves6 records

Expansion highlights7 records

GibsonAI competitors and assessment

Company assessment

Direct peers

  • Mem0: Mem0 is a direct competitor offering a memory layer for LLM applications and AI agents, with automatic memory extraction, storage and retrieval across sessions. It targets the same developer persona, uses a similar self-serve/SDK model and competes head-to-head on accuracy benchmarks and token-cost reduction.
  • Zep (GetZep): Zep provides long-term memory and a temporal knowledge graph for AI assistants and agents, with APIs for message storage, summarization and fact extraction. It overlaps closely with Memori on agent memory infrastructure and on the use of structured recall rather than raw context stuffing.
  • Letta: Letta (formerly MemGPT) builds persistent memory and stateful agent infrastructure for LLM applications. It competes directly on long-horizon agent memory, context management and developer-facing APIs for production deployments.
  • Cognee: Cognee offers an open-source memory layer for AI applications that structures and retrieves knowledge from conversations and documents using graph and vector storage. It is comparable to Memori on the agent-memory abstraction, structured recall and self-hostable architecture.

Broad incumbents

  • LangChain: LangChain is the dominant LLM application framework with its own agent orchestration, memory abstractions and integrations across hundreds of vector stores. It is a broad incumbent whose native memory components overlap with Memori's value proposition for the same developer audience.
  • LlamaIndex: LlamaIndex provides data frameworks for LLM applications including indexing, retrieval and agent memory components. It is a broad incumbent in the same ecosystem, offering overlapping capabilities around structured context and agent-side recall.
  • Pinecone: Pinecone is a leading managed vector database used for retrieval-augmented generation and agent memory use cases. It is a broad incumbent with significantly more capital and a larger installed base that could expand into the agent-memory category Memori occupies.
  • Weaviate: Weaviate is an open-source vector database with hybrid search, modules and a managed cloud service widely used for RAG and agent memory. It is comparable on the retrieval and semantic-search backbone that Memori layers its agent-memory product on top of.

Emerging players

  • Chroma: Chroma is an open-source embedding database popular with AI developers for RAG and lightweight agent memory. It overlaps with Memori on the developer-targeted self-serve distribution and on serving as a retrieval layer underneath agent applications.
  • Qdrant: Qdrant is an open-source vector similarity search engine with a managed cloud offering, used for semantic search and memory in LLM applications. It is comparable as a retrieval infrastructure layer that Memori's customers and competitors rely on for agent context.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat5 records

Key risks6 records

Key highlights6 records

Customer concentration

GibsonAI social profiles

Digital presence

GibsonAI compliance and trust

Trust signal

Compliance2 records

GibsonAI financial estimates

Financial estimate

Revenue estimate

Valuation estimate

GibsonAI leadership team

Management profile

Number of profiles

GibsonAI funding detail

Funding detail

Funding overview

Funding rounds1 record

Investors4 records

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

GibsonAI M&A and investment

M&A and investment

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about GibsonAI

What does GibsonAI do?

GibsonAI (operating as Memori Labs Inc., DBA Memori) provides an LLM-agnostic memory infrastructure layer that turns AI agent execution and conversation into structured, persistent state for production systems. Its product captures each chat turn, classifies content into facts, preferences, rules, and summaries, and serves targeted recall with semantic search to reduce LLM token costs. The offering includes a Python SDK for drop-in integration and a hosted Memori Cloud service.

Is GibsonAI a public or private company?

GibsonAI is a private company. It is classified as venture growth investor backed and is currently operating.

When was GibsonAI founded?

GibsonAI was founded in 2024. It employs 1 to 10 people.

Where is GibsonAI based?

GibsonAI is headquartered in San Francisco, United States, in the North America region.

How does GibsonAI make money?

Two revenue lines are on record. Memori Cloud Subscription is the primary driver. The others are self-serve SDK.

Who are GibsonAI's main competitors?

Direct peers on record are Mem0, Zep (GetZep), Letta and Cognee. Broad incumbents are LangChain, LlamaIndex, Pinecone and Weaviate. Emerging players are Chroma and Qdrant.

Does GibsonAI have an API?

Yes. Drop the SDK into existing code to handle model calls and callbacks with zero configuration. The SDK enables developers to integrate Memori's memory infrastructure into their AI agent workflows. Developer documentation is at memorilabs.ai/docs.

What industry is GibsonAI in?

GibsonAI's product category is AI Memory Infrastructure. Its primary akta.pro industry code is HDAEANAD, LLMOps & Generative AI Platforms (Prompt/Agent Orchestration, RAG), with a secondary code of HDAAAAAF, AI Memory & Storage for Training (HBM/DDR/SSD/Object Storage). Its NAICS code is 518 and its SIC code is 7372.

Unlock the full company data

50 free credits on sign-up, no credit card required.

Contact sales
Live signals
Arizona Daily StarBusinesses worry about perception of Tucson mass shootingBusinesses along East Congress Street in Tucson kept doors closed after Sunday's mass shooting left 10 people injured, including the suspect. The police tape remained for most of the day, and owners said they did not grasp the impact until the next morning.American City Business JournalsGibson CEO Cesar Gueikian steps downGibson Inc. CEO Cesar Gueikian is stepping down. The announcement provides no further details on the reason or timing. No successor or replacement is named.MarkTechPostGibsonAI Releases Memori: An Open-Source SQL-Native Memory Engine for AI AgentsGibsonAI released Memori, an open-source SQL-native memory engine for AI agents that stores memory in standard databases like SQLite or PostgreSQL. It claims 80-90% lower infrastructure costs and 10-50ms query times compared to vector databases. The system enables portable, auditable memory for any LLM.GitHubGibsonAI/Memori: Open-Source Memory Engine for LLMs, AI Agents & Multi-Agent SystemsMemori is an open-source SQL-native memory engine designed for large language models (LLMs) that enables persistent and queryable memory through a simple integration. The tool claims to offer cost savings by eliminating the need for expensive vector databases and supports multiple SQL databases like SQLite and PostgreSQL. A private beta for Memori v3 is being established for early access to its new features.Packt HubGibsonAI Memori: SQL-Native Memory for Agents, NVIDIA’s Universal Deep Research, Conversational Commerce Agent on Vertex AIThis article is a weekly newsletter summarizing various updates in data science and AI, including the launch of NVIDIA's Universal Deep Research prototype and Google Cloud's general availability of GKE Inference Gateway and Conversational Commerce agent. It also highlights GibsonAI's release of Memori, an open-source SQL-native memory engine for AI agents, alongside tutorials on federated learning security and Python-based economic modeling.