Sergey KimanAI Product Engineer

End-to-end AI products, Python backends, data pipelines, Telegram experiences, and web interfaces.

Sergey KimanAI Product Engineer

End-to-end product · Telegram + Web

F-Signal

Marketplace intelligence delivered through Telegram and a Mini App.

I designed and built the product end to end: product logic, system architecture, Python backend, data and analysis pipeline, authenticated API, Telegram delivery, WebApp, validation, and release boundaries.

Green location pin with a check mark on a transparent acrylic panel
Public F-Signal product artwork. Product details are documented in the sanitized case.

What it does

F-Signal monitors fresh Kufar listings, normalizes product facts, compares prices with market baselines, performs AI-assisted risk analysis, matches opportunities to purchasing profiles, and delivers selected signals through Telegram and a Mini App.

System shape

  1. Listing ingestion
  2. Normalization
  3. Market baseline
  4. Structured analysis
  5. Scoring
  6. Profile matching
  7. Durable signal state
  8. Telegram and WebApp delivery

Key engineering decisions

A staged pipeline

Each transformation has an explicit responsibility instead of mixing collection, analysis, ranking, and delivery.

Shared durable signal state

Telegram and WebApp consumers read the same persisted product state.

Protected operational boundary

Authenticated product APIs and operator tooling remain separated from the public surface.

Structured delivery system

The project was developed through my own structured delivery system: documentation, decisions, validation gates, release and rollback discipline, and separation between public product surfaces and private operations.

Result and evidence

The Hmara-hosted production MVP reached an owner-provided pipeline snapshot dated 2026-07-17, reviewed for this case: 2,993 stored listings, 5,444 seen listing IDs, 2,494 evaluations, 3,385 AI analysis runs with 3,258 successful, 2,215 unique listings analyzed by AI, 370 fetch runs, 555 Telegram notifications, and 21 test files. These figures describe pipeline volume and engineering maturity, not paid traction, revenue, customer growth, or established demand.

Scope boundaries

  • The implemented first vertical focuses on iPhone marketplace listings.
  • The private source repository, operational details, provider configuration, and user data are not publication artifacts.
  • No paid-customer, revenue, demand, or scale claim is made.

Open to remote product engineering roles

I am looking for a stable remote role where I can own meaningful product or engineering outcomes across AI-assisted systems, Python, automation, integrations, Telegram, and web products.

Choose how to reach me

Both options go directly to Sergey. Pick the channel that fits your conversation.