Kirill

Context Engineer

I ship real, production software by directing AI end-to-end — context, memory, and orchestration, not autocomplete. Every project below, including this site, was built that way.

See the work Get in touch

I build complete products — solo — by orchestrating Claude across every stage of the work: research, planning, multi-file implementation, and verification. That's not a shortcut; it's a real, deep skill — knowing what context an agent needs, how to structure it, when to plan before executing, and how to check results instead of trusting them blindly.

Six shipped products back that up: a commerce platform, a full casino game provider with its own remote gaming server, a real-money Telegram Mini App backend, an AI assistant running inside a live Telegram Business account, a production shop bot, and a handful of polished client sites — spanning web apps, bots, and game engines.

Selected Work

Shipped, not prototyped

Six case studies. Client and gambling-adjacent projects are shown as screenshots — no live links, so this site doesn't send traffic to someone else's brand or a money-gaming site.

Vitrea — screenshot pending

Vitrea — Digital Goods Store

Full-stack commerce platform built solo: Next.js 16 + Prisma/PostgreSQL storefront selling licenses, keys, subscriptions and templates in 10 languages, with full RTL support for Arabic. Every key fulfillment runs inside a database transaction with a claim-and-retry guard — a digital key can never be sold twice, even under concurrent checkouts. Two-tier admin (Owner/Staff) with masked customer emails, CSV export, and a built-in, cookieless analytics dashboard. Payments sit behind a pluggable provider interface, ready to swap in a live processor without touching order logic.

Next.js 16 Prisma PostgreSQL Tailwind v4 10 locales

View live ↗

Casino Games — screenshot pending

Casino Games & RGS Provider

A self-built casino game provider running five live games — two slots (Zeus vs Hades, Book of Awakening), European roulette, crash, and baccarat — on one shared architecture: a headless, seeded math engine verified with a Monte Carlo simulator over millions of rounds per game, served through a signed, SQLite-backed remote gaming server. Table games land exactly on their textbook numbers — roulette resolves to the analytic 36/37 RTP on every bet type. Two of the five run on fully custom Three.js 3D tables (PBR materials, a physically-animated ball, card-dealing choreography), wired end-to-end to the real engine and live money, not just a visual demo.

PixiJS 8 Three.js Vite Custom RGS

Casino Platform — screenshot pending

Casino Platform

Real-money-grade backend for a Telegram Mini App casino, hardened by a 7-detector fraud engine — RTP exploitation, net-extraction, multi-accounting by IP, bonus abuse, ledger-integrity breaks — that can auto-freeze withdrawals platform-wide. Every balance mutation runs under row-level locking with idempotency keys on every external-facing action. Beyond the wallet: a four-mechanic bonus system with real wagering-requirement tracking, a streamer referral program paid on net GGR (excluding house-issued money, so a bonus can't be farmed for commission), live-scored tournaments, a battle pass, and genuine Telegram Business API gift redemption — real NFT gifts, reconciled by a background worker so a network hiccup can never double-pay. A separate admin panel runs its own Telegram-authenticated SPA with a full audit trail.

Node.js PostgreSQL TON Connect Telegram Business API Fraud detection

PiarBot — mockup pending

PiarBot — AI Business Assistant

An AI assistant that runs inside a real Telegram Business account, drafting replies to rental clients and sending only on manual approval — never automatically. Structured LLM output drives an 8-stage CRM pipeline, voice-note transcription, and rate-limited, debounced batching to keep cost predictable. A second layer mines every conversation for product complaints, classifies them deterministically into 7 categories, and rolls them into a prioritized roadmap for the business owner.

Python aiogram 3 OpenAI Telegram Business API

Seller-04-bot — mockup pending

Seller-04-bot — Telegram Shop

A Telegram storefront handling real orders through a payment processor with no per-transaction references — reconciling payments by amount/currency/status and using a time-boxed lock to stop two buyers colliding on the same tracked payment. Later extended to accept orders from a companion web storefront through the same order table.

Python aiogram PostgreSQL APScheduler

Client Sites — screenshot pending

Client Websites

A set of zero-build-step static sites for individual clients: a bilingual, 8-language, RTL-aware personal/business site with a hand-built glass UI (blur, SVG noise-grain, mouse-tracked spotlight) and a full legal public-offer page with a scroll-tracked table of contents; plus several lightweight petition and donation pages with live signature counters. Built for clients who needed something polished with zero ongoing maintenance burden.

HTML / CSS / JS Alpine.js i18n RTL

Process

How it's built

Every project here starts with a written brief — a CLAUDE.md — and a persistent memory of decisions, constraints, and past mistakes that carries across sessions, so context isn't rebuilt from scratch every time. Non-trivial work goes through a planning pass before any code is touched, and research fans out across parallel agents when the scope is broad, instead of guessing. The result is software that's actually finished — the process is built to catch gaps before they ship, not after.

Brief

A CLAUDE.md documents the goal, constraints, and decisions for each project before any code is written.

Memory

A persistent, cross-session memory carries context and open items forward — nothing gets rebuilt from scratch.

Plan

Non-trivial changes get a written plan and sign-off before implementation starts.

Verify

Claims get checked against the real code — simulations run, numbers measured, not assumed.

Skills

Two tracks, one workflow

AI orchestration

Context engineering CLAUDE.md authoring Persistent memory design Multi-agent workflows Planning before execution Verification-first process

Technical

Next.js / React / TypeScript Node.js Python PostgreSQL / Prisma PixiJS / Three.js Telegram Bot & Business API

Let's talk

Open to full-time roles and freelance work — especially anything involving AI-directed development, context engineering, or shipping real products fast.

Telegram WhatsApp — coming soon