From Request to First Meeting: How United Mentors Evolved into a Comprehensive Platform
Until recently, most mentorship services solved only one problem: providing a directory of experts and a way to submit a request. Everything that followed—refining the inquiry, introductions, scheduling, correspondence, documentation, and payment—was left entirely to the client and mentor.
We have completely rebuilt this journey at United Mentors. It is no longer just a storefront of profiles, but a full-fledged platform where one can find a suitable mentor, get introduced, schedule meetings, formalize the partnership, and continue working—without the need for manual contact transfers or fragmented messaging.
The main change isn't a single new feature. We have integrated every stage of mentorship into a unified, intuitive process and automated the tasks that previously required constant intervention from a manager.
AI matching starts with a task, not filters
Entrepreneurs rarely need just a "marketing mentor" or a "retail expert." Usually, the request is more complex: the company has reached a certain scale, old channels have stopped working, the team is stalling, and the owner cannot identify the real bottleneck.
That is why we built an AI agent that helps formulate the request itself. It clarifies business context, current status, required competencies, and the results the entrepreneur wants to achieve. The dialogue is concise: the agent asks one meaningful question at a time and does not force the user to fill out a long questionnaire before demonstrating value.
Once sufficient information is gathered, the system matches the request against published mentor profiles. It considers not only the general industry but also the business stage, the mentor's experience, their competencies, and the specific challenges they solve. In the candidate's card, the client can see why this particular person was recommended—for example, due to experience in a specific field, work with a similar business stage, or relevant management expertise.
Matching can begin even before registration. If a user selects a mentor, the dialogue and results are saved: after logging in, they are transferred to the personal dashboard, where the user can proceed and send a request.
This is a key principle for us: AI should not replace human decision-making. Its task is to eliminate the need for the client to manually study dozens of profiles and to shorten the path to a few truly relevant candidates.
Two dashboards, one shared process
We created separate dashboards for clients and mentors. Each side has its own interface and actions, but both are connected by a single scenario.
The client dashboard stores the history of AI matching, suggested candidates, submitted requests, and the current status of work with each mentor. Here, users can select an introductory time, view upcoming meetings, go to a chat, request an invoice, begin mentorship, and sign an NDA if necessary.
In their dashboard, the mentor receives new requests, views the client's formulated task, accepts or rejects the introduction, suggests available times, and manages active dialogues. This space also includes notification settings, the profile, platform operational data, and a referral section.
The entire process has become transparent. A request no longer disappears into a form with an ambiguous status. Both sides see the sequence of stages: request, scheduling, introductory meeting, final decision, payment, and the start of regular sessions.
Before payment, a 30-minute introductory meeting takes place. This allows the client to verify not just the mentor's resume, but also personal compatibility—how easy it is to communicate, whether the mentor understands the context, and if their way of thinking is a good fit. After the meeting, the decision rests with both parties; mentorship does not start automatically just because a client selected a profile.
Communication remains within the platform
One of the most noticeable problems in recent months is messenger fragmentation. Some continue to use Telegram, others have moved to MAX, while some prefer not to conduct work conversations in personal accounts. Because of this, even simple communication between two people can be interrupted at the most inconvenient moment.
We decided not to tether mentorship to a specific messenger. Once work begins, the client and mentor have access to a direct chat within United Mentors. It stores the unified history of correspondence, system events, meeting details, and documents. If one participant changes their preferred channel, the shared dialogue does not disappear.
In addition to direct communication, the dashboard includes separate dialogues with the AI agent and United Mentors support. This separates different contexts: matching questions are not mixed with work agreements, and inquiries to the platform team are not lost among the mentor's messages.
At the same time, we maintained the convenience of Telegram. It can be connected as an additional transport layer: new messages arrive in a bot, and the response from the bot is returned to the same dialogue on the website. For example, a mentor can reply from Telegram, and the client can reply from their dashboard. They continue to communicate even though they are using different interfaces.
Personal contact information does not need to be exchanged. The bot does not show the counterparty's phone number or account, and the canonical history remains on the platform. This protects the privacy of both parties and prevents important agreements from dissolving into personal correspondence.
Meetings are scheduled without endless back-and-forth
Organizing the first meeting often takes more time than the meeting itself: one person sends several time slots, the other replies a day later, and by then, some options are already taken.
Now, a mentor can suggest available times in a simple message—for example: "Wednesday from 2:00 PM to 4:00 PM" or "I'm free Friday after 11:00 AM." The AI recognizes the proposal, accounts for time zones, and turns it into clear 30-minute slots. If the date or time is ambiguous, the agent asks for clarification rather than creating a meeting at random.
The client selects a suitable option in the dashboard. After confirmation, the platform automatically creates a conference in Yandex Telemost and adds the link to the meeting. It does not need to be created, copied, or sent separately to the second participant.
We also implemented a unified notification layer. The client and mentor receive meeting confirmations, followed by reminders one hour and five minutes before the start. The dashboard features unread message indicators and sound notifications; if an important message remains unanswered for a long time, the system can gently provide a reminder.
Automation here is not for the sake of automation itself. It frees people from organizational work so they can get to the substantive conversation faster.
Contracts, payments, and NDAs are now built into the workflow
Mentorship involves strategy, finances, team matters, and other sensitive business information. Therefore, the legal aspect should not exist separately from the product in the form of random files and emails.
For the client, a public offer applies on the platform. When setting up their profile, the mentor selects their working structure—sole proprietorship (IP), self-employed, legal entity, or non-resident—and confirms the corresponding cooperation terms with United Mentors. For sole proprietors, the self-employed, and companies, agency contracts are provided: the platform acts as the bridge between the client's payment and the mentor's work.
After both sides decide to proceed, the client can request an invoice or pay via bank details. The payment status is linked to the specific partnership and displayed in the shared process, so participants do not need to manually check if funds were received or if the work chat is active.
After payment, the client can propose a bilateral NDA directly in the chat. The mentor confirms their details and formally accepts the agreement. The system records the document version and the acceptance of both parties, and the finalized NDA remains linked to the specific mentorship.
An NDA remains voluntary: not every request requires one. However, if the conversation involves trade secrets, deals, new technology, or internal reporting, confidentiality can be established in the same place where the work is conducted, without exchanging files via email.
The platform helps, but it doesn't remove the human from the process
We have automated matching, scheduling, notifications, and part of the document flow, but we are not building a service where people are left alone with an algorithm.
The United Mentors team can see the path of each pair and can step in if the matching proved difficult, if the mentor hasn't replied for a long time, if a question about payment arises, or if the participants need help. The client can write to support from their dashboard, and a manager will respond in that same dialogue. If necessary, manual curation supplements the AI results.
For us, this is the correct role of technology in mentorship: not to substitute trust between people, but to remove the friction surrounding it.
What else is new in United Mentors
Along with the core scenario, we added several important features around the platform.
First, a referral system is now live. Mentors can invite new clients and mentor candidates, use personal links, and see the status of recommendations and earnings in their dashboard. For external partners, we also created a separate page with program rules.
Second, we established a transparent mechanism for charitable donations. A mentor can voluntarily choose a foundation and donate 5, 10, 15 percent, or their entire income from the mentorship. The donation is calculated only from the mentor's fee, does not affect the platform commission, and actual transfers and payment documents are published on a separate page.
Third, we strengthened internal support: we consolidated the history of inquiries, application statuses, payments, and meetings, added notification delivery tracking, and included system signals for situations where the process has stalled. The user does not see most of this infrastructure—and that is exactly how it should work. Its task is to ensure that no agreement is lost.
Not a mentor directory, but an environment for collaboration
Ultimately, United Mentors has become an intermediary between people, technology, and various communication channels. A client no longer needs to search for an expert separately, explain their request to multiple managers, move communication to a personal messenger, manually coordinate times, create a video call, and compile documents.
In turn, a mentor does not need to build their own sales funnel, administer every introduction, and keep track of dozens of organizational details. They receive a formulated request and can focus on what the client came for: analyzing the business, asking powerful questions, and finding solutions.
We wanted to make the path to a mentor shorter and make mentorship itself more stable and secure. Now, the entire chain, from the first phrase to the AI agent to regular work meetings, is truly gathered in one place.