Frame the real problem
Start from the repeated friction, the existing workflow and the evidence that software should exist at all.
Independent AI Software Studio
Ideas into
useful software.
AI agents, automation, developer tools and focused software systems — designed, built and documented by one independent builder.
ModeIndependent
FocusAI-native systems
MethodSmallest reliable system
OutputSoftware + build notes
Currently building
Software that coordinates software work.
Current experiments focus on long-running agent workflows, durable project state and practical automation that turns repetitive development work into reusable systems.
Explore DeepLoopMinder
Selected work
Projects
DeepLoopMinder
An AI-native multi-agent software development runtime focused on project control, durable state and external coding workers.
IssuePack
A local-first context packaging tool that turns messy customer chats, screenshots and attachments into traceable issue packages for coding agents.
AutoScreenAgent
An experimental Android agent that turns natural-language goals into screen actions through accessibility APIs and LLM tool calling.
How I work
From repeated friction to a system.
The goal is not to add AI everywhere. It is to identify where software can remove repeated effort, then make that software explicit, observable and maintainable.
Prototype the smallest system
Prove the workflow with the fewest moving parts before adding orchestration, persistence or model complexity.
Make state explicit
Keep context, decisions and progress observable so long-running software can be resumed, inspected and trusted.
Harden what survives
Turn the useful prototype into maintainable software with clear boundaries, tests, documentation and automation.
What I build
Services
AI Agent Systems
Design and implementation of practical agent workflows, tool use, task routing and durable project state.
Automation & Internal Tools
Replace repetitive operational work with focused automations, data pipelines and small tools that remain understandable.
Web Systems
Fast, maintainable websites and focused web applications with clean information architecture and strong technical foundations.
Current focus
Building the infrastructure around AI work.
Most current experiments sit one layer below the chatbot interface: orchestration, context, durable state and the operational tooling that makes model-driven work dependable.
Agent coordination
Long-running agent workflows, delegation and durable execution without losing project state.
Context infrastructure
Reusable issue packages, structured evidence and compact context that survives across tools and sessions.
Operational automation
Replacing repetitive software work with observable systems that remain understandable to a human operator.
Notes from the lab
Build in public.
Why this studio site is static-first
The public Yachenyi site treats crawlability, speed and machine-readable structure as architecture constraints rather than launch-time polish.
Agent workflows need durable project state
Long-running software work becomes more reliable when agents operate on explicit projects, tasks and artifacts instead of a single disappearing chat context.
Have a project in mind?