Available for new work

Frontend, and the data infrastructure underneath it.

I build interfaces that render live data — and the scrapers, schedulers and pipelines that keep them fed. Most recently: a production engine that monitors 1,300+ assets around the clock and writes decisions back through an API, unattended.

Start a conversation See selected work
1,300+
assets tracked continuously in production
24/7
unattended operation, no manual intervention
Solo
architecture through deployment and ops
01

What I build

Two halves of the same job: the surface people look at, and the machinery that makes the numbers on it true.

Dashboards & product UI

Next.js and TypeScript frontends over whatever backend you already have. Server components, real-time updates, fast on mobile, correct SEO.

Market data & pipelines

Ingestion from external sources, normalisation, scheduled workers with retry, backoff and rate-limit handling. The unglamorous layer that everything else depends on.

Telegram Mini Apps

Full stack — the interface, the bot, the backend and the deployment. Shipped as one piece, not handed over in fragments.

02

Selected work

The domain below is commerce, not crypto — I'd rather say that plainly than dress it up. What transfers is the engineering, and I've noted exactly where.

Flagship In production

Real-Time Competitive Pricing Engine

Solo engineer — architecture, implementation, deployment, operations

Problem

An operator with 1,300+ active listings was losing sales to competitors who repriced faster. Manual monitoring doesn't scale past a few dozen items — by the time a price was checked by hand, it was already stale.

What I built

  • A server-side engine continuously polling competitor listings on an external marketplace
  • A normalisation layer mapping external listings onto internal catalogue entities
  • A rule-based decision engine writing adjusted prices back through the platform's REST API
  • Resilient scheduling — retry with backoff, rate-limit and anti-blocking handling, failure alerting
Result: reaction time went from sporadic and manual to continuous. The system has run unattended in production since deployment.
Why this matters for crypto

Structurally, this is a price-oracle feeder. Poll external market data → normalise → apply rules → act. Point it at a DEX or exchange endpoint instead of a marketplace and the architecture is unchanged. Same problem, different asset class.

PythonREST integrationScheduled workersServer deployment
FrontendMigration

Headless Storefront — Next.js 16 over a Custom JSON API

Solo engineer — full stack

Problem

The client's catalogue was locked inside a legacy PHP CMS: slow, weak on mobile, and effectively impossible to iterate on.

What I built

  • A custom JSON API exposing the legacy CMS catalogue as clean endpoints
  • A Next.js 16 frontend consuming it — server components, image optimisation, correct canonicals and pagination
  • Migration executed without breaking existing URLs or search rankings
Result: a modern, fast storefront over an unchanged legacy backend — no risky data migration required.
Why this matters for crypto

Every protocol has a backend it can't replace and a frontend it wants to modernise. This is that job, already done once.

Next.js 16ReactTypeScriptRESTLegacy integration
ETLAutomation

Automated Feed Pipeline — 1,300+ Entities

Solo engineer

What I built

An automated pipeline generating a spec-compliant XML feed for 1,300+ catalogue entities, kept in sync with the source of truth and validated against a strict schema. I also built and ran the paid acquisition campaigns consuming it.

Why it was hard

  • Schema validation with zero tolerance — one malformed entity rejects the batch
  • Scheduled regeneration kept consistent with a live, changing catalogue
  • Multi-currency campaign structure on top of the generated data
Result: the full catalogue exposed to paid channels with zero manual feed work.
Why this matters for crypto

Scheduled ETL over 1,300+ entities with schema validation is the same shape as indexing on-chain data into an analytics dashboard.

XML / ETLScheduled generationMerchant CenterGoogle Ads
03

Stack

Frontend

Next.js 16 · React · TypeScript · Tailwind · server components · edge deployment

Backend & data

Node · Python · REST integration · scheduled workers · scraping & normalisation · ETL

Infra

Cloudflare Pages · VPS & shared hosting · cron orchestration · monitoring & alerting

Also

Telegram Bot API & Mini Apps · Google Merchant Center · headless CMS integration

Got frontend or data work you'd rather not hire for?

I take on project work directly, and white-label subcontracting for agencies — under your brand, your process, your client relationship. Currently have capacity.

Message on Telegram Email me
@stesous vladburakov540@gmail.com