Dmitry K.
Offer Dmitry work on your next project.
Private message
Offer work
Ukraine
14 days 22 hours back
available for hire
on the service 1 year
Rating
Successful projects
Average rating
Skills and abilities
Programming
- AI & Machine Learning
- Bot Development
- Data Parsing
- Databases & SQL
-
Desktop Apps
from 23 USD for project
- HTML & CSS
-
Javascript and Typescript
from 11 USD for day
-
Python
from 11 USD for day
-
Web Programming
from 16 USD for project
Services
Administration
Portfolio
-
1400 USD Store in Telegram bot
Bot Development
The bot supported a service catalog, order placement, chat between the client and the administrator, several payment methods, and a workflow for approving the result before final delivery.
It implemented manual payment through administrator confirmation, payment with cryptocurrency, and Telegram Stars. The system supported prepayment, previewing the result with an automatic watermark, and issuing the final file without a watermark after confirmation and payment.
… The bot also included a two-way client-admin chat: both parties could open a dialogue, while closing the chat remained the administrator's responsibility. Text messages, photos, and videos were transmitted through the bot, allowing it to be used not only as a showcase for services but also as a full-fledged working channel for order management.
The project underwent a complete rework: initially, the bot was focused on selling in-game currency, and later it was redesigned into a media material ordering system. This provided experience not only in developing a Telegram bot but also in restructuring the product logic for a new usage scenario.
-
999 USD KasAgent — desktop AI agent with a smart model dispatcher
Desktop Apps
KasAgent was created as a personal "neural network operator" for Windows: not just a chat, but an agent that can read a project, plan steps, and execute them through the operating system's tools.
For working with LLM, I integrated KasAgent with the existing OmniRoute proxy and MCP server (this is a separate open-source/external component, not my development). The agent uses this layer to dynamically access different models and providers, but the logic for selecting models and orchestrating tasks is implemented on my side — in the Intelligent Model Dispatcher.
… The agent operates in several modes: Assistant (read-only, no write access), Executor (full access within permissions), and a sub-mode Planning UI, which shows the user a plan of steps before execution. In a separate Fast mode, one can quickly run an idea or check without heavy planning when complex orchestration is not needed.
For project work, there is a multi-project workspace: the agent knows the root directory PROJECTSDIR, sees several repositories, can perform listdir, search for files, read source code, and save artifacts of its work in ~/.kasagent/artifacts (text, HTML, JSON, etc.). All operations through file tools undergo path validation and are limited to the working directory, so the agent cannot "wander" throughout the system.
A separate artifact system is implemented: the agent can save intermediate reports, compressed summaries, code snippets, and visual results as files to return to them later. For complex tasks, there is support for context compression and "thought signature" — a separate brief summary of the model's thoughts that can be mixed into subsequent requests instead of the full log.
Through the browser part (Chromium in Electron), KasAgent can orchestrate interface checks: take screenshots, run them through vision models, confirm text from the screen, and based on this, generate a report on the test result — this scenario is currently shown in the screenshot.
5–7 bullets "what KasAgent can do":
Two-phase architecture: the Intelligent Model Dispatcher I wrote plans the task, validates models, and hands off execution to the KasAgent Executor.
Integration with the external OmniRoute proxy and MCP server: KasAgent decides which models to request through this layer but is not the author of OmniRoute.
Multi-project workspace: the agent sees several projects, reads files, searches through code, and saves work artifacts in ~/.kasagent/artifacts.
Strict permission system: separate modes for reading and writing, path validation, and restriction of all operations to the working directory.
Vision tools: describe_image, reading text from screenshots, and generating human-readable reports on the results of interface tests.
UI on Electron + React with a separate Planning UI window, where the agent first shows the action plan and only then starts execution.
Support for quick hypothesis testing through Fast mode without heavy planning when response time is important.
#electronjs #Node.js #javascript #React #desktop-applications #Linux\Windows #ai-integration
-
1500 USD KasHost — personal hosting with an admin panel and API node
Javascript and Typescript
Own hosting panel / PaaS system for remote project and node management. The admin panel works like a remote router: it allows switching between multiple nodes on different devices (PC, Termux on mobile devices, etc.) and managing their projects from a single interface.
Key functionality:
… - Support for multiple nodes: switching between devices with an installed node, each with its own set of projects and configurations.
- Full process management: starting, stopping, restarting, viewing statuses and logs for each project.
- File manager: remote management of the file system of devices with a node (viewing, uploading, deleting, editing) — available as a separate tab in the side menu, as well as a file block on the specific project page.
- Remote node updates: updating directly from the panel with auto-backup before applying. Simply upload an archive with the new version — the node updates and you can always roll back.
- Security and production readiness: user account system, user tokens, and master tokens for access management.
- Special logic for Minecraft servers: working with the plugin list, auto-installation via Modrinth API and other services, editor for server.properties and service configs.
- Support for different types of projects: managing dependencies directly from the panel (installing/removing), configuring the project path, launch commands, and project type.
- Isolation modes: no isolation, isolation via venv for Python projects, planned support for containerization (Docker/PRoot) depending on the platform.
Architecture features:
- Cross-platform node: tested on Windows, Linux (Ubuntu/VPS), and Termux on Android. The project was initially designed with the aim of running on different devices — from home PCs to remote servers and mobile nodes.
- Special local authorization mode: in the local environment, you can connect to the node from the panel without a master token, using a simplified local token, which is sufficient for local use. For remote nodes, master and user tokens are used.
- Extensibility: the panel was originally conceived as a foundation for public hosting — the architecture considers the addition of new project types and deployment scenarios. In the current version, it is used as private hosting.
The project has been developing for about a year: it initially started as an internal panel for managing a single node, and over time was reworked into a cross-platform PaaS system with support for multiple nodes, security, remote updates, and specialized scenarios (for example, for Minecraft servers).
#backend #nodejs #typescript #docker #paas #adminpanel #api #devops
