Back to projects
AUTOMATION / CASE STUDYCode reviewed

Ofertas 24HFrom discovery to distribution.

A deals and affiliate platform with an analysis engine, price history and job-based publishing.

Node.jsJavaScriptSQLiteTelegram APICSV / XML

From discovery to distribution.

A deals and affiliate platform with an analysis engine, price history and job-based publishing.

My role

Product definition and AI-assisted development, connecting requirements, implementation and review.

Local code and README reviewed: Node.js, SQLite, worker, moderation services, analytics and providers. Continuous operation and external integrations were not tested in this review.

Deals arrive in different formats. Publishing without normalization, history and duplicate control creates noise and unreliable discounts.

A modular pipeline separates collection, normalization, evaluation, moderation and distribution. A scheduler coordinates jobs while SQLite stores products, history and clicks.

Providers Documented manual, CSV and authorized feed inputs. Integration code does not imply active external access.

An explanatory diagram of the documented scope. This is not live system monitoring.

0–100 Deal Score and deduplication

Implemented¹

Price history and moderation

Implemented¹

ManualProvider, CSV and generic feeds

Implemented¹

Scheduler, jobs and cooldown

Implemented¹

Tracked redirects and analytics

Implemented¹

Store integrations dependent on access

In development

¹ Identified in code or implementation documentation; not equivalent to a new production test.

Node.jsJavaScriptSQLiteTelegram APICSV / XML

Distinguishing real discounts from artificial changes, preventing duplicate posts, enforcing cooldowns and handling each affiliate integration’s requirements.

Actual screenshot of the Ofertas 24H public showcase, under preparation

Public showcase under preparation; it does not represent the local dashboard or prove bot operation.

Separating providers from the deal engine lets data sources evolve independently of publishing rules. History and moderation are part of data quality.

KEEP EXPLORING

AI Knowledge System
LET’S BUILD SOMETHING.

An interesting problem?
Let’s build.

Web systems, automation and tools that turn an idea into working software.