Skip to content
View kalinichenko88's full-sized avatar
🏠
work from home
🏠
work from home
21 followers · 53 following

Achievements

Achievement: Pull Sharkx3Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: YOLOAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pull Sharkx3Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: YOLOAchievement: Arctic Code Vault Contributor

Organizations

@getsedona @sedona-cms

Block or report kalinichenko88

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
kalinichenko88/README.md

About Me

I design and build production systems around AI agents—and the infrastructure they run on. My focus includes multi-agent runtimes, MCP servers, long-term memory layers, model routing and failover, and Telegram and Discord integrations.

I bring 15+ years of production experience across frontend and full-stack development. I currently work as a Senior Frontend Engineer at Apliteni, building frontend systems for a large self-hosted product.

What I Work On

  • AI infrastructure and agent systems — Agent runtimes, MCP servers, memory layers, model routing, failover, and chat integrations
  • 0→1 full-stack products — Backend, frontend, billing, email, privacy, infrastructure, and documentation
  • Frontend modernization and platforms — Legacy migrations, design systems, component libraries, testing, and build infrastructure

Selected Projects

  • 🤖 Telegram Agent Kit — Runtime-agnostic ESM library for wiring LLM agents to Telegram with a zero-dependency core, Markdown-to-HTML rendering, live draft streaming, and a turn-loop bridge
  • 📚 VaultMD — Headless markdown-vault data layer for Bun with CRUD over .md notes, a derived SQLite index, backlinks, and keyword search
  • 🗞️ AI Digest — Automated tech news digest powered by Claude Code and MCP that collects, deduplicates, and summarizes RSS feeds and GitHub releases
  • 💰 Obsidian Budget Planner — Minimalist budget planning plugin for Obsidian, built around markdown code blocks inside notes

Tech Stack

AI & Agent Systems

LLM agents · MCP · Claude Code · Agent memory · Model routing · Telegram Bot API

Backend & Data

Node.js, Bun, NestJS, PostgreSQL, SQLite, Redis, and GraphQL

Frontend

React, TypeScript, JavaScript, Vue, Nuxt, Next.js, Astro, HTML, CSS, Sass, and Tailwind CSS

Tools & Infrastructure

Git, Docker, GitHub, GitHub Actions, VS Code, Neovim, Webpack, Vite, Vitest, and Markdown

Pinned Loading

  1. obsidian-budget-planner-plugin obsidian-budget-planner-plugin Public

    A minimalist budget planning tool for Obsidian. Create, track, and manage budgets using markdown code blocks directly in your notes.

    TypeScript 11

  2. ai-digest ai-digest Public

    Automated daily tech news digest powered by Claude Code + MCP. Collects RSS feeds and GitHub releases, deduplicates, and summarizes into a personalized Markdown digest.

    TypeScript 1

  3. vaultmd vaultmd Public

    Headless markdown-vault data layer for Bun — CRUD over .md notes with a derived SQLite index for collection queries, backlinks, and keyword search.

    TypeScript

  4. telegram-agent-kit telegram-agent-kit Public

    Runtime-agnostic ESM library that wires LLM agents to Telegram — zero core dependencies. Markdown→HTML rendering, live draft streaming, and a turn-loop bridge over injectable Bot API interfaces.

    TypeScript

  5. sedona-cms/blocks sedona-cms/blocks Public

    Block editor for Sedona CMS

    TypeScript 1 1

  6. constlab/vesper-jwt-auth constlab/vesper-jwt-auth Public

    TypeScript 1