EXMEAgent
Manager

Your personal AI assistant

Tell us
what you need.EXME gets
to work.

From your first request to a finished result.
Models, tools and a workspace are already brought together in one assistant.

See how it worksRead the whitepaper
Your requestPrepare a client proposal
EXME
ResultProposal · structure and figures
Easy to start. Room to do more. You stay in control.The product, its engine and its economics

01 / FROM REQUEST TO RESULT

Your task.
Its job.

Describe your goal in everyday language. Your assistant brings together the context, the right tools and the next steps.

Example request

“Compare three options. Show me which one suits me and why.”

An informed choice

01
Clarifies your needsBudget, timing and what matters most.
02
Gathers evidenceSearching, reading material and comparing sources.
03
Prepares the resultA comparison table and a conclusion with caveats.

These examples illustrate the workflow. An assistant’s capabilities depend on its connected tools and the permissions you grant.

02 / WHY EXME

An agent’s capabilities.
Without another
job to learn.

01

Start with the task

The platform maintains the workspace and model connections. You do not need to deploy a server or assemble an agent stack.

02

Pick up where you left off

Memory, files and context help you return to your work. Clarify important requirements and revise information that is out of date.

03

Grant just the access needed

Connected services extend your assistant’s capabilities. Access to personal accounts and consequential actions stay under your control.

04

See what the work costs

Model usage tracking and spending limits help you manage consumption. Convenience rests on measurable economics.

THE EXME ECOSYSTEM

Planned offering

Hundreds of plugins.
Thousands of skills.
Out of the box.

One assistant for many kinds of work. We plan to bring connections to your work services and ready-made task workflows together in EXME.

01 / PLUGINS

A connection to your systems.

A plugin connects your assistant to an external service, helping it retrieve data, work with files and prepare actions in familiar tools.

02 / SKILLS

A ready-made way to get it done.

A skill defines a workflow: how to conduct research, assemble a document or check a result. You just describe what you want to achieve.

  • Email
  • Documents
  • CRM
  • Analytics
  • Development

Personal services connect with your permission. You choose which data and actions your assistant can access.

Hundreds of plugins and thousands of skills describe the target scale of a future offering. This is a development plan, not a verified count of connections available today.

FROM REQUEST TO RESULT

Behind one request,
a clear route.

Follow a task through instructions, model selection and a connected tool. Each stage has a distinct role.

  1. 01

    Your request

    Prepare a short summary of a document.

  2. 02

    Skill

    Instructions: how to review the material and format the answer.

  3. 03

    Router

    Selects a route for the task’s requirements.

  4. 04

    LLM model

    Budget route

    A short task with limited context.

    Complex route

    Multiple sources and more reasoning.

  5. 05

    Plugin

    Tool: retrieve a document through a service API.

  6. 06

    Model: verification

    The tool response returns to the model for review and the next step.

  7. 07

    Result

    A summary with the key findings for you.

An illustration of the principle · 7 stages. Choose a scenario and start the walkthrough.

Savings become margin. In this planning example, costs fall from $6 to $3 per customer with 75% of comparable workload on an economical model. This is a hypothesis, not measured savings.

Check the assumptions

The diagram illustrates task-based model selection. Routing is implemented; realized savings and quality require measurement. Model availability depends on the validated connection.

A skill provides instructions; a plugin provides a tool or API. Personal services connect with your permission, and consequential actions remain under your control.

NEXT STAGE / PLANNED

Chain crystallization
into pipelines.

First, the agent works out how to complete a task. Then a validated action chain can be “baked” into a reusable automation.

01 / DISCOVERSolve and validate

The agent finds a working sequence and checks the result.

02 / CRYSTALLIZEBake the chain

Fix the steps, inputs, permissions and acceptance checks in a versioned pipeline.

03 / REPEATRun at lower cost

Routine runs follow the pipeline without new model reasoning at every step. Exceptions return to the agent.

Planned capability. Potentially fewer model calls, lower cost per repeat task and higher margins. API, infrastructure, monitoring and maintenance costs remain; this additional saving is not included in the financial scenarios.

03 / UNDER THE HOOD

One place to start.
A complete system inside.

The engine connects models, memory and tools. EXME manages access, the working environment and costs.

01 / INPUTA familiar chat

Your task, materials and permissions.

02 / EXECUTIONThe EXME engine

Context, memory, a model and a sequence of actions.

03 / RESULTFinished work

A document, research or an approved action.

Control panel · access and spendingPersonal workspace · files and toolsLLM providers · connected services

A managed working environment lets users focus on the result. Data flows through EXME, model providers and connected services according to the task and granted permissions.

Explore the engine

04 / MODELS AND DEVELOPMENT

Models evolve.
Your assistant stays with you.

The provider registry allows the model lineup to grow. Each connection is checked for quality, tool support and cost accounting.

GLM

Initial configuration

A budget-model family included in the current product configuration.

Claude

Model support

The architecture supports the family; availability depends on validated connections and the package.

More providers

Expansion path

Compatible APIs and connection presets provide a path to DeepSeek, Kimi, MiniMax and OpenRouter.

Selection by task

Cost optimization

Model routing is implemented; its quality and savings are being evaluated for user scenarios.

PLANNED PRICING

Choose your level of use.

Planned prices, not a public offer. Included usage, limits and package details will be validated in the pilot. The model lets you compare price points and adjust delivery and acquisition costs.

05 / ECONOMICS

Growth you
can calculate.

Explore the model
RevenueAll operating costs
Target monthly price$25per customer
First operating profitMonth 3growth target
Cumulative paybackMonth 6startup costs and losses covered
Peak funding need$14,605including startup costs

Growth target at $25/month: 220 new paying customers per month, 3% monthly churn, $12 CAC, $3 AI, $1 infrastructure and $1 support per average paying customer; fixed costs $6,000/month. This requires validated low-cost acquisition channels, retention and measured delivery costs. These are targets, not achieved results or a return guarantee. Taxes and financing excluded. Compare with the base case in the model.

GROWTH / SCENARIO

A path to scale

Illustrative growth case: 65% on $25 and 35% on $50; ARPU $33.75. Starts with 300 new paying customers per month, growing 12% monthly. CAC starts at $25 and grows 2% monthly; churn is 3%. Premium delivery costs are 1.8× standard costs. The mix and churn are held constant across the customer base.

Paying customers · end of month 2429,394
MRR · end of month 24$992.1k
ARR · annualized run rate$11.9m
Peak funding need$33.2k

ARR is 12 × end-of-month MRR, not revenue earned during a year. Modelled revenue in year 2 is $6,256,495. Revenue in month 24 is $936,672, based on average customers; its operating profit is $504,828, before tax and financing.

Operating setup: 1–2 people for support and owned servers. The scenario allocates $6,000/month to the team and other fixed costs; no additional per-customer support fee is assumed. Infrastructure starts at $1 per standard customer for power, connectivity, maintenance and a replacement reserve; premium usage raises this budget. This is an illustrative cost budget, not verified capacity.

The growth path depends on acquisition economics, retention and capacity validation. At this customer count, measure concurrent task load, server headroom, support requests and response times. Add expansion costs or reduce growth if current hardware and 1–2 support staff cannot meet service levels. Pipeline crystallization savings are not included.

Explore the growth case

06 / NEXT STEPS

Progress through
proven results.

Each stage ends with a review of quality and economics. Capabilities expand as the product becomes ready.

01

Repeatable value

Select useful scenarios and make the journey from setup to result clear and reliable.

Criterion: task completed and accepted
02

Sustainable economics

Measure delivery costs, retention and acquisition payback across paying cohorts.

Criterion: positive contribution
03

More capabilities

Expand models, skills and integrations. Scale infrastructure to match actual demand.

Criterion: quality holds as usage grows

07 / CLEAR ANSWERS

What to know
before you start.

Do I need to understand the setup?

EXME maintains the platform’s infrastructure and connections. To work with your email or another personal service, you need to grant access. The assistant cannot authorize itself on your behalf.

Can the agent handle any task?

Capabilities depend on the model, tools and permissions. The assistant can make mistakes, so review important results. Consequential actions need appropriate oversight.

Where is my data processed?

Data may pass through the chat platform, EXME control panel, assistant environment, LLM and connected services. The whitepaper explains the architecture and processing boundaries.

Is $25 the current public price?

This is the value in the reviewed test deployment and the starting assumption for the financial model. Commercial access terms, included usage and limits will be specified when the offer launches.

Where can I find the calculations and plans?

The whitepaper explains the product and its architecture. The financial model lets you change assumptions, compare three scenarios and download all 24 months of calculations.

Behind a simple request,
a carefully built system.

Open the whitepaper