INVESTORS & PARTNERS
An AI operating system,
before it is obvious.
DRENTIR is a pre-MVP working prototype: an AI operating layer that turns business requests into verified work. This page is the short version — the full materials are shared in conversation.
Not a pitch deck — an introduction. No confidential materials are published here.
Problem
AI answers. Work stays manual.
Businesses drown in coordination: inboxes, scheduling, follow-ups, reports. Chatbots generate text; the operational work still belongs to humans.
Product
An operating layer for agents.
DRENTIR takes requests in plain language and coordinates agents, tools and models to complete them — with permissions, human approval and audit built in.
Why now
Models are ready. Control was missing.
Capable models — including local ones — finally exist. What was missing is the control layer that lets AI act inside permissions: identity, policy, verification, audit. That is what DRENTIR builds.
INITIAL MARKET
Voice and operations agents for small businesses.
Not ten industries at once. The first job to be done: handle incoming requests, prepare and follow up on communications, coordinate calendar, web and operational tasks.
01
Handle incoming requests
Calls, messages, emails — triaged, understood, routed.
02
Prepare communications
Drafts, briefs and follow-ups — ready for human approval.
03
Coordinate operations
Calendar, web and recurring tasks — executed inside permissions.
PLATFORM ADVANTAGE
Control is the moat.
- Local + cloud routing: the right model for every subtask — privacy and cost by design.
- Permissions below the model: tenant identity is injected below the LLM layer — the model never decides its own reach.
- Human approval and audit trail as primitives — not add-ons.
- One platform, many verticals: the same core powers business agents and the Tuteshko city guide.
CURRENT STAGE
Pre-MVP. Working prototype. Honest numbers.
DRENTIR is a pre-MVP AI operating system in active prototype development. Core platform components are already working internally; end-to-end integration, validation and external pilots are in progress.
40%
Overall MVP readiness
70%
Core Platform
65%
Identity & Permissions
65%
Model Routing & Local AI
Readiness estimates reflect product integration and validation status, not lines of code. Last updated: October 9, 2026.
Flagship proof
Tuteshko — an AI city guide built on DRENTIR
Voice, RAG, maps, location, route planning, mobile app, analytics, model routing — one vertical application proving the platform is real.
In Integration
SEEKING
Four doors in.
01
Private pilot partners
Small businesses ready to delegate real operations and shape the product with feedback.
02
Strategic partners
Distribution, domain expertise, and deployment channels for the first verticals.
03
Technical partners
Model providers, hardware, and infrastructure — for local AI at real scale.
04
Investment conversations
We are open to conversations with investors who understand platforms are built before they are obvious.