Shared Business Knowledge
Documents, data, decisions and operating rules become usable context, so every task starts informed instead of from zero.
CREATIVE TECHNOLOGIST
Ten years building digital worlds for film, games and fashion. Today I build the AI agents and systems that run applications, workflows and operations. Same obsession, new tools.
WHAT I BUILD
I design connected agent systems that use your knowledge, work through your existing tools and carry real work from request to result.
Documents, data, decisions and operating rules become usable context, so every task starts informed instead of from zero.
One agent or a coordinated team can research, reason and plan around a goal, involving people whenever judgment matters.
Agents work through the software you already use: email, CRM, databases, APIs and internal systems.
Repeatable work moves from request to completion with fewer handoffs, less duplication and clear checkpoints.
Dashboards, portals and conversational interfaces give people a simple way to direct, review and use the system.
Permissions, activity and outcomes stay visible, while successful work becomes reusable knowledge for what comes next.
ALSO IN THE TOOLBOX
Image and video generation systems for brands and creators, including adult content. Self-hosted models, custom LoRA training, and consent, rights and moderation tooling you control.
Deployment, monitoring and maintenance of always-on agents on your own servers, with secrets, logs and permissions handled properly.
Agents that live inside Telegram, WhatsApp and Slack, plus content pipelines that draft, schedule and publish across social channels.
CHATBOT
AGENT
Conversation is the interface. Execution is the point.
THE OPERATING MODEL
An operating model where one shared agent holds the company's context, and humans and specialist agents pick up each other's work without meetings, handoffs or waiting.
Traditional work organizes time. Remote work moves the same structure to another room. I build operations around a shared agent instead.
At the center sits one agent: the company terminal. It holds the knowledge, the decisions and the work in progress, and every person and every specialist agent connects to it directly. No bottlenecks, no waiting for someone's calendar.
Because the agent is the shared memory, work is continuous. A human can pick up where an agent stopped, an agent can continue what a human started, and new people are onboarded by the system itself instead of by weeks of alignment meetings.
In this model, technical background stops being the gate. Clear goals, planning and problem solving are what create value, and they create it at a speed traditional structures cannot match.
DEVELOPER PROJECTS
Choose which AI models do the work, build the workflow around them, and run it with a predictable cost per model. One graph your whole team can run, review and reuse.
Once a flow works, it is saved as a graph, not as one person's memory. Anyone on the team can open it, run it and improve it.
Bring your own API keys. You pay providers directly, no markup, no seats, full visibility on spend.
Pipelines can be triggered by an agent and pushed straight into Pabbly automations that schedule and publish. Generation becomes an operation, not a task.
Custom LoRA, upscaling, compositing and export nodes. Made for creative teams who need control over the result, not a button to press.
The goal is not to turn creatives into button operators. It is to let them iterate faster and control the outcome better.
THE JOURNEY
From film frames to virtual worlds to AI systems: the medium changed, the obsession didn't.
THE CRAFT
Virtual fashion, world builds, post-production for film and advertising. Made one polygon at a time, long before any of it could be generated.
CONTACT