Home/Personal Brain
Personal Brain · a new W2W product
A personal AI assistant that knows your business
Not Siri and not a voice assistant that opens YouTube. Personal Brain lives in your Telegram, remembers every meeting, note and agreement — and answers in seconds, because your data lives in Obsidian and is indexed in Qdrant.
- $5,000fixed turnkey price
- 2 weeksto a working assistant
- Telegramone chat, text and voice
- 100%your data, in Markdown
What did we promise Aziz at Thursday’s meeting? And remind me about it tomorrow at 10
- You promised to send the October volume by October 1.
- He gives a 7% discount with a 50% prepayment.
- Open question — lead time for the 5,000-unit run.
How it is different
Not a voice assistant. One person’s second brain
Siri, Alexa and Google Assistant know nothing about your business. Personal Brain is built on your data and works like a personal assistant who sat next to you in every meeting.
- Opens YouTube, sets a timer, plays music
- Remembers nothing about you between conversations
- Doesn’t know your clients, prices or agreements
- Answers from the internet, not from your documents
- Doesn’t do work tasks
- Remembers every meeting, note, voice message and chat
- Answers from your data with a link to the source
- Knows your clients, suppliers, prices and who promised what
- Sets tasks, reminds you and writes drafts on its own
- Sends the day’s plan in the morning and a summary in the evening
Who it’s for
For those who keep the whole company in their head
1Small business owner
A shop, a workshop, services, a chain of outlets. Dozens of contacts and agreements a day — and no system where any of it is written down.
2Expert and consultant
Lawyer, doctor, financier, realtor. Clients, cases, documents — the assistant keeps the context for each one.
3Founder and executive
Meetings, investors, team, product. Decisions stop getting lost between calls.
4Blogger and author
Ideas, drafts, content plan, partners. The assistant keeps an idea bank and reminds you when it’s time to publish.
What it does
Six skills in one chat
Everything through Telegram: by text or voice. Nothing to install, nothing to learn.
Remembers everything
Forward a voice message, a note, a photo of a document, a link — the assistant parses it and files it in the knowledge base. A month later it finds it by meaning, not by keyword.
Answers from your data
“How much did we pay for rent in March?”, “What did the client from Samarkand ask for?” — an answer in seconds, with a note on where it knows that from.
Runs your meetings
A recording or a voice note after the meeting → summary, decisions, who promised what, tasks with deadlines.
Sets tasks and reminds you
“Remind me a day before the payment”, “Assign a task to Bakhtiyor” — tasks go to Google Tasks or your tracker, the reminder arrives in Telegram.
Writes drafts
A reply to a client, a proposal, a post, a letter to a supplier — in your style, with facts from your base. You only edit and send.
Morning and evening digest
At 8:00 — the day’s plan, meetings, who needs a reply. At 20:00 — what got done, what is stuck, what’s tomorrow. Without you asking.
How it works
Why it is fast and why the data stays with you
Three parts: Telegram as the interface, Obsidian as a knowledge store in an open format, Qdrant as semantic vector search. No black box.
Telegram
The only interface. Text, voice messages, photos, documents, forwarded messages.
Assistant core
A language model with skills: understands the request, searches the base, performs the action, answers with a link.
Obsidian
A knowledge base in plain Markdown files. Opens on your phone and laptop, readable without us.
Qdrant finds the right note among thousands in a fraction of a second — the answer arrives faster than you could find it yourself.
Markdown in Obsidian means your files. If you leave us, the base stays and keeps opening.
We deploy Qdrant on your server or in the W2W cloud. Only your Telegram account has access to the bot.
Personal Brain in action
One day of a business owner with the assistant
No new apps and no new habits: you write and talk in Telegram as usual — the assistant does the rest.
Morning digest
“3 meetings today. At 11:00 — Aziz, you promised him the October volume. Two clients have been waiting for a reply since yesterday. The rent invoice is due the day after tomorrow.”
A voice note after the meeting
You dictated 2 minutes in the car. The assistant wrote a summary, pulled out 3 agreements and created the task “send the specification by Friday”.
A question on the go
“What price did we quote for this same product last year?” — an answer in 3 seconds with a link to last year’s proposal.
A draft reply to a client
You forwarded the client’s message — and got a ready reply in your tone with the current terms. Changed one word, sent it.
Evening summary
“5 tasks closed, 2 moved. Tomorrow: reply to Aziz about lead times, pay the rent. New notes in the base: 7.”
Plans
What your Personal Brain costs
Two plans with a fixed rollout price plus a small monthly fee for the model, server and support. We fix the scope in a proposal after a 30-minute call.
$5,000
One owner. Memory, search, meetings, tasks and integrations with your business.
- Telegram bot: text, voice messages, files
- Knowledge base in Obsidian with no volume limit
- Qdrant vector search in the W2W cloud
- Summaries of meetings and voice messages, automatic transcription of calls and Google Meet
- Tasks and reminders (Google Tasks / Calendar)
- Morning and evening digest
- 3 systems connected: CRM, email, Google Drive / Sheets
- Drafts of proposals, letters and posts in your style
- Access for an assistant or deputy (2 accounts)
- Custom scenarios: up to 3 skills for your process
- Live in 2 weeks
Support: from $150 / month (model, server, updates)
$10,000
Everything in Start plus a private environment, voice, a team and unlimited skills.
- Everything in Start
- Qdrant and the bot on your server: data never leaves your perimeter
- Unlimited integrations: CRM, 1C, email, telephony, spreadsheets
- Voice mode: calls and dictation
- Up to 5 users with different permissions (family, assistants, partners)
- Unlimited skills for your process
- Personal manager and SLA
- Upgrade to Company Brain as the team grows, without losing the base
Support: from $300 / month · under SLA
from $15,000
An assistant for a specific profession or a family office. Your own environment, your own skills.
- Everything in Max
- Fully private installation on your servers
- Unlimited integrations and skills
- Several users with different permissions
- Personal manager and SLA
Support: under SLA
The rollout price is fixed and does not change after the start. The monthly fee depends on the request volume and the chosen model.
How implementation works
Two weeks later you are talking to your own assistant
1Call and collection
We go through what you do, where the data comes from and what annoys you every day. We collect the first notes, contacts and documents.
Result: a map of sources and scenarios
2Build
We set up the Obsidian base, Qdrant and the Telegram bot. We load your data and tune the answer style and digests.
Result: the assistant answers your questions
3Test drive
You use it for a week as usual. We watch where it makes mistakes and fine-tune the skills and integrations.
Result: 30 real questions, an 85% threshold of acceptable answers
4Handover
We show how the base is organised, how to add data and how to edit it. We hand over the access and the manual.
Result: Personal Brain at work, support by subscription
What you get
Four guarantees in the contract
Acceptance threshold
30 of your real questions. Below the threshold — we fix it at our own expense.
The data is yours
Markdown in Obsidian and an export from Qdrant. If you leave, you take everything.
From the first call
Your clients, prices and correspondence never leave the project perimeter.
Start after the contract
The team starts right away; the prototype is ready in two weeks.
Questions about Personal Brain
If the business grows and the team needs the assistant, Personal Brain upgrades to Company Brain — the same knowledge base plus access rights, employees, CRM and 1C.
How is this different from ChatGPT in Telegram?
ChatGPT knows nothing about your business and forgets the conversation. Personal Brain is built on your knowledge base: notes, meetings, documents, correspondence. It answers from your data with a link to the source, remembers the context for months and takes action: sets tasks, reminds you, writes drafts.
What is Obsidian and why do I need it?
Obsidian is a free note-taking app that stores everything in plain Markdown files on your device. The assistant writes and reads exactly these files. You always see what it knows, can open the base on your phone and fix things by hand. And the base does not depend on us.
What is Qdrant and why are the answers fast?
Qdrant is a vector database. It stores the meaning of every note and finds the right one among thousands in milliseconds, even if you asked in different words. So the assistant does not “re-read” the whole base for every question — it picks the 3–5 relevant fragments straight away.
Where is my data stored?
Notes — in your Obsidian (on your device and in your cloud). The vector index — in the W2W cloud or on your server, your choice. The language model only receives the fragments needed for the answer. We sign an NDA before the first call.
Can the assistant make up an answer?
Every statement comes with a link to a source in your base. If confidence is below the threshold, the assistant says so: “this is not in the base”. At acceptance we check 30 real questions against an 85% threshold.
Do I need to install anything?
No. The interface is the regular Telegram you already have. Obsidian is optional, for browsing the base yourself.
What does the monthly fee include?
The language model, the server for Qdrant and the bot, skill updates and support. From $150 a month on the Start plan; the exact amount depends on the request volume.
What if the business grows and the team needs the assistant?
Personal Brain upgrades to Company Brain: the same knowledge base plus access rights, employees, CRM and 1C. Nothing you built is lost.
We reply within an hour during business hours
Let’s discuss your project
Two short steps. You can skip the second one — we’ll discuss the details on a call.
- 130-minute callWe analyze the task and propose a solution
- 2Proposal in 2–3 daysScope, timeline and budget fixed before the start
- 3Prototype in 2 weeksCompany Brain already running on your data
Nuriddin AbdumalikovCOO — personally responsible for kicking off your project