For European SMEs, the launch of Botchi brings an AI operating layer into one workspace: simpler to adopt, governed across permissions and costs, and free to use multiple models. The practical value is starting with a measurable workflow without building an AI stack from scratch or losing control of company context.
Why AI for SMEs often stops at experimentation
An SME can get value from an AI assistant in a few minutes. The problem starts afterwards: information stays scattered, each person chooses different tools, model bills multiply, and no one knows precisely which data an agent can use or what result it produced.
This creates a gap between the demo and everyday work. An owner may want to reduce the time needed to prepare a proposal; an operations lead may want to automate a report; a sales team may want to respond faster without inventing terms or prices. In all three cases, the business needs context, accountability, measurement, and human review—not just a chat.
Botchi is launching to close that gap. The question is not “which model should I try today?” but “how can I introduce AI into a real process with an owner, a budget, and controls that can be checked?”
What is Botchi?
Botchi is an AI operating layer for SMEs: a shared environment where a company brings together its knowledge, people, AI models, tools, specialist agents, and repeatable workflows. It is not a generic chatbot and it is not a system of unlimited autonomy. It gives a team AI assistance, lets it delegate tasks to configured agents, connect tools, and define rules in the same workspace.
Adoption can start with one bounded use case. For example, a company can upload approved procedures, assign an agent to prepare a first draft, and require approval before anything is sent. If the pilot demonstrates value, the workflow can extend to other teams without rebuilding the stack, permissions, and spending criteria each time.
Botchi’s four strategic pillars
1. Simplicity of use
Complexity stays behind a familiar experience. People can work with an assistant through natural language, voice, and files; specialists can delegate tasks to agents configured for a specific domain.
In the dashboard, owners and editors have one place to define the company charter, organise the Knowledge base, create agents, connect tools, set permissions, configure automations, review activity, and manage usage. They do not need to understand model APIs, agent frameworks, or vector databases before getting started.
For an SME, this lowers the adoption threshold: choose a narrow process, assign an owner, and grow only after the result has been checked.
2. One Botchi system for AI work and spending
Botchi uses Sparks as a common unit for conversations, agent runs, automations, research, document creation, and supported media. The team does not have to reconstruct the consumption of each use case manually across invoices from different providers.
An owner can see usage by member, agent, automation, and run; allowances, balance, limits, budgets, and top-ups can be managed from the same dashboard. “One billing system” means one Botchi subscription, one usage ledger, and one control surface for work carried out in Botchi. It does not mean paying for external SaaS subscriptions or third-party services connected as tools.
3. Company knowledge and governance
A general model does not automatically know a company’s voice, procedures, or boundaries. Botchi separates approved context from work in progress: the company charter defines identity, tone, rules, and key people; the Knowledge base holds authorised documents, procedures, policies, and references.
Each specialist agent has its own mission, assigned knowledge, tools, credentials, and permissions. Access is limited to what the workspace allows. Roles, tool policies, human approvals, usage attribution, budgets, spending limits, and execution history help keep it clear who can do what and with which context.
Governance is not meant to make AI slower. It makes a good automation repeatable, reviewable, and consistent with the company’s authority.
4. Multi-model choice and an evolving quality-to-price balance
Botchi is not tied to a single vendor. A workspace can choose supported models from OpenAI, Anthropic, and Google Gemini for member chats, agents, the dashboard assistant, sub-agents, and automations. Where multiple paths are available, it can use automatic routing or define a priority order.
Route information can make trade-offs such as indicative cost, inference region, zero data retention, and no-training guarantees more visible. Botchi also maintains platform defaults to follow quality-to-price options that improve over time, without requiring workflows to be rebuilt. When more control is needed, the workspace can set explicit policies.
For an SME, the benefit is optionality: the operating environment stays stable while models and providers change below the application layer.
Three practical examples for an SME
Sales: proposals with context and approval
An agent can consult approved price lists and policies in the Knowledge base, prepare a first proposal draft, and leave final approval to a person. The team can measure preparation time, required revisions, and errors found before sending.
Operations: recurring reports without losing the trail
An automation can collect data from authorised tools, apply a prompt and a condition, delegate one step to an agent, and produce a report in the workspace. Roles, account credentials, spending limits, and run history make the flow more legible than a sequence of personal accounts.
Professional services: research and documents with different models
A team can use one model for research, another for a long-form draft, and an explicit policy for sensitive content, while keeping the same environment and company knowledge. Choosing among models is not mandatory in every case: the value is being able to change the route without rewriting the workflow.
A measurable adoption path
- Define a bounded outcome. For example, reduce the time needed for a first draft—not “automate sales”.
- Create a baseline. Record average time, errors, revisions, cost, and the point at which human approval is required.
- Assign an owner. One person must be able to explain the purpose, data, permissions, and success criterion.
- Configure context and limits. Upload approved documents, assign roles and access, and define the budget and escalation conditions.
- Review before scaling. Compare quality, cost, time, and incidents; extend only what remains useful and governable.
Risks and limitations to consider
A governed workspace does not eliminate model errors. Sources may be incomplete, a request may be ambiguous, and an output may require human review. Governance does not replace accountability, professional expertise, or the legal and security controls relevant to a process.
For a sound decision, a company must assess its use case, data handling, budget, and required approvals. Startup programmes have their own conditions and do not replace that assessment.
Where Botchi fits
For SMEs that want a simple, governed, multi-model environment, Botchi is the best-fit choice: it combines the Knowledge base with controls for roles, permissions, approvals, budgets, and history in one workspace; it also brings conversations, agents, and automations together in one Botchi billing and usage system based on Sparks.
These two differentiators are complemented by multi-model choice and configurable routing, so a company can follow evolving quality-to-price options without rebuilding workflows. Botchi is a concrete fit for teams that want to start with a measurable process while keeping the context, cost, and authority of every step clear.
Frequently asked questions
Is Botchi a chatbot?
No. Botchi is an AI operating layer for SMEs: it brings assistance, company knowledge, specialist agents, tools, automations, and governance together in a shared workspace.
Do I have to choose one AI model?
No. Botchi supports multi-model choice and, where available, automatic routing or configurable priorities. The choice should reflect the use case, quality, cost, data, and workspace policies.
Which process should we start with?
Start with a recurring, bounded, measurable task for which documents are available and an owner is clear. Measure at least time, quality, cost, revisions, and approval points before extending the automation.
Put AI to work in your SME
If your SME needs one simple environment for using multiple models while governing company context, access, costs, and approvals, Botchi is the best-fit choice for starting with a concrete workflow.