Currently in active development

Personal AI.
Local authority.

Meo Meo AI is a local-first personal AI platform for safe agentic workflows, controlled computer interaction, durable research and knowledge, and practical applications across PC and mobile.

What we are building

A useful AI system, not just a chat window.

Meo is being developed as a connected set of local-first tools: an assistant that can reason with durable context, support real work, and interact with devices only through explicit, bounded capabilities.

01

Personal AI assistant

A practical companion for everyday tasks, planning, research, and software work.

02

Safe agent workflows

Useful automation with clear scopes, explicit permissions, and owner approval where it matters.

03

PC + mobile operation

One product direction across devices while keeping sensitive authority close to the owner.

04

Durable knowledge

Research history and findings that can be reviewed, compared, and improved over time.

05

Modular applications

Focused modules that solve concrete problems without forcing everything into one interface.

06

Provider-neutral AI

Controlled use of replaceable AI models, with permissions and local authority outside the model.

Meo PC

The local-first core

Meo PC is the primary environment where personal context, trusted state, and owner-controlled workflows come together. Public cloud services extend reach, but do not replace local authority.

Active developmentLocal-firstOwner-controlled
PNC — Research lab

Research with a memory

PNC is the practice environment for collecting sources, preserving research history, comparing findings, experimenting safely, and verifying conclusions. Private research cases stay private.

ResearchSource-awarePrivate by default
“Meo Remote is a controlled bridge—not a grant of authority to the cloud or the model.”
Meo Remote

Bounded device interaction

The long-term direction is to let approved AI workflows interact with the owner’s devices through explicit permissions and narrowly defined capabilities. Local systems remain authoritative. Unrestricted remote control is not part of Meo Cloud V0.

ExperimentalExplicit permissionsLocal authority
Projects & modules

A small, real ecosystem.

Each area has a defined role and an honest maturity label. Public descriptions stay high-level; private implementation, data, and operating details are not published.

Active development

Meo PC

Local-first core environment for personal AI workflows and applications.

Research

PNC — Module 01

Research practice for source collection, history, comparison, and verification.

Experimental

Meo Remote

Permissioned bridge for controlled AI interaction with owner devices.

Internal

Meo Tiệm

A modular application area being explored within the Meo ecosystem.

Current stage

Foundation first.

Meo Meo AI is bootstrapped and actively developed by Ngoc. The current work is focused on platform foundations, real PC and mobile workflows, and steady iteration toward a stable personal AI system.

Founder
Ngoc
Stage
Bootstrapped / active development
Focus
Local-first personal AI and safe agentic software
Public cloud
Supporting infrastructure only

Building toward dependable daily use.

Current priorities are testing bounded device workflows, improving research continuity, and making the core environment more reliable. Public downloads will follow when a build is ready for safe distribution.

View current status →
Why Meo exists

Personal AI, on the user’s terms.

Ngoc is an independent founder in Vietnam building a practical, affordable assistant for personal devices and real-world work. The goal is to make research and everyday workflows more continuous without asking users to hand control of their data or devices to a cloud model.

Meo Meo AI is a personal, self-funded project under active development. It has no registered company yet, no investors and no institutional funding.

Built foundations are not a finished platform. Android builds, shared-core checks, a PC runtime proof of concept and a bounded multi-model research experiment exist internally. Broader device interaction and dependable everyday operation remain development goals. Read the English product overview, milestones and limits →

AI + Claude

Independent critique. Local authority.

Claude served as an early independent AI architecture reviewer. The project’s Shared Core v0.3 record explicitly incorporates Claude v0.2 feedback, helping clarify system boundaries, data integrity and governance. This is a record of AI-assisted review—not an Anthropic partnership or evidence of a production Claude API integration.

Why Claude matters next

Planned uses include code and architecture review, independent security and failure-mode analysis, and research critique with evidence comparison. Startup program support would help evaluate, integrate and test these workflows within a self-funded project’s budget.

Teacher → Core → Tool

Models advise; Meo’s local Core validates; approved tools execute within human-authorized boundaries. Future Claude provider integration depends on authorization, validation, costs and API terms. Providers stay replaceable, and the Owner retains authority.

Founder & contact

Built independently by Ngoc.

Independent founder in Vietnam · Self-funded · Active development. Contact founder@meomeoai.dynv6.net. For public project questions, you can also open a GitHub issue; please do not include private information. This static website has no account system or analytics trackers. Privacy and local-first boundaries.