Product & design
Shaped the product, designed the interface, and built workflows for entrepreneurs running websites, funnels, courses, and customer conversations
Open to product & AI engineering roles
I build AI products from idea to production. As the solo founder and engineer behind OctoFunnel, I connect full-stack engineering, product design, and an understanding of how customers run their businesses. The figure is combined customer sales through OctoFunnel funnels since 2022
Bali, Indonesia · UTC+8 · Open to relocation

Customer figures are approximate and based on platform records. Sales belong to customers; they cover the platform’s history
The supplied database-check summary reports 2,087,071 verified contact records across 187 client installations, excluding our own platforms.
Approximately 5 million is the upper end of the supplied 3–5 million lifetime estimate, not a verified total. The estimate allows for closed or unreachable client sites, including assumed audiences for 387 historical sites with recorded sales whose contact counts could not be measured.
Records include leads, subscribers, and customers. A person can appear on several sites, and some contacts were imported. These counts do not establish unique people, paying customers, simultaneous active users, or contacts generated by the AI agent.
Source: the owner's supplied database-check screenshot. The underlying calculation and queries were not included in the attachment, and this workspace has not independently rerun the count. View the supplied source summary
The product
A platform for digital entrepreneurs to build websites and sales funnels, deliver courses, and manage customer conversations. I brought those workflows together in one product and added an AI agent that can operate the same workspace through chat
Explore all platform featuresShaped the product, designed the interface, and built workflows for entrepreneurs running websites, funnels, courses, and customer conversations
Built and operated the platform end to end. Developed the legacy PHP/MySQL system and the current TypeScript, React, and Node.js product
Designed a shared virtual filesystem for the built-in AI agent and MCP-connected agents, with contextual instructions, validated changes, checkpoints, and durable execution
Developed the commercial model and a network of around 30 sales partners, using regular partner conversations to identify product needs
Platform capabilities
I built connected tools for creating the product, finding customers, delivering the offer, and running the day-to-day business. AI assistance and visual editing bring those workflows into one workspace
One connected
business workspace
Explore the platform · 16 feature areas
Turn a business idea into structured pages, content, and workflows through conversation
Create and publish the customer-facing home for a business
Connect messages, pages, decisions, and offers into a customer journey
Give the AI useful source material about the customer’s audience, offer, and expertise
Keep customer records, conversations, and funnel progress connected
Let customers enter and continue the funnel in familiar messaging apps
Create conversational assistants that talk to the business’s customers
Reach an audience and manage ongoing membership in one workspace
Deliver learning content and keep students moving through the experience
Connect an offer to payment records and the access a customer buys
Manage the products and variants behind a small online shop
Offer calls and services with a configurable appointment workflow
Create the visuals used across pages, campaigns, and learning content
Prepare and produce the content behind a lesson or social campaign
Organize the people, responsibilities, and material needed to run the business
Connect marketing signals, business activity, and partner programs
Engineering case study
My key decision was to expose the platform’s business resources as a virtual filesystem. Both the built-in AI agent and agents connected through MCP inspect and change the same structured resources. Customers continue editing that data visually
The shared interface combines nine filesystem operations, contextual playbooks, and server-side validation. Checkpoints and durable execution help handle changes and interruption
Combined activity from the built-in AI agent and MCP-connected agents working through the same virtual filesystem. Verified from retained production records on 9 October 2026; calls include reading, navigation, and changes
24,706 built-in agent calls + 14,886 MCP calls = 39,592 recorded calls across 115 projects. 35,380 returned a successful technical result (89.36%); 8,608 of 11,954 write, edit, copy, move, or delete attempts succeeded (72.01%). Calls can repeat or make no effective change. 113 owner accounts used chat or MCP; 111 had recorded tool activity. The provisional cohort excludes configured operators and test-like accounts and may include staff-assisted activity. Logs can be incomplete or deleted. Customer acceptance, time saved, and cost per accepted task remain unmeasured. The original report records seven passing and two failing targeted tests
Read the evidence and source referencesCustomer outcome
Vitaly Okunev · platform customer
~$3.75M
Customer sales through the platform · approximate USD equivalent
I built the platform and funnel infrastructure the customer used. Its configured journey connects Telegram signup, four lesson pages, timed follow-up, homework, CRM, and a paid course. The customer brought the offer, audience, and business execution
Approximately $3.75M in USD equivalent, using a fixed reference conversion consistent with the reported customer case. Contact and purchase counts were checked in the client database on 9 October 2026. This case contributes to the approximately $32M combined customer-sales total

The person behind the product
I’m looking to bring that ownership to a team building useful AI products. I enjoy turning a complicated workflow into something people can understand and use, working close to customers, and taking responsibility for the result
I started building this platform in 2019 and grew it through a partner network. The business continues to serve customers. I’m now seeking an engineering role where I can contribute that experience to a team and a new product challenge
I have worked with AI APIs since 2023 and have two years of experience building AI agents. Claude Code and Codex are part of my development workflow, alongside design, implementation review, and testing
Read my résuméStarted building the product to support an online education business
Developed the business and partner-led distribution under the MetaFunnels name
Introduced the current brand and AI-oriented TypeScript platform
Tools I work with
Legacy platform: PHP, MySQL, JavaScript, and jQuery
Let’s build something useful
Open to AI Product Engineer, Product Engineer, and Founding Engineer opportunities
kozyonicon@gmail.comBased in Bali, Indonesia · UTC+8
Remote work; open to relocation
English: C1
Meeting window: 07:00–20:00 Bali
This is my scheduling window. Working hours and overlap can be agreed for the role