Upwork home
Sign up
Project Catalog
DesignDesign
Video & AudioVideo & Audio
Development & ITDevelopment & IT
MarketingMarketing
Writing & TranslationWriting & Translation
MoreMore

You will get a broken Python web scraper fixed, optimized, and QA-validated

Andrii V.Status: Offline
Andrii V. Andrii V.
Rising Talent
Preview image

Let a pro handle the details

Buy Data Mining & Web Scraping services from Andrii, priced and ready to go.
Andrii V.Status: Offline
Andrii V. Andrii V.
Rising Talent
Preview image

Let a pro handle the details

Buy Data Mining & Web Scraping services from Andrii, priced and ready to go.

Project details

⚠️ Please MESSAGE me with your script + target URL before buying, so I can give you a free feasibility check first.

Your Python scraper broke and you're losing data. I fix it, harden it, and validate the output against your real data — so it actually works, not just runs.

I diagnose and fix:
• Layout/HTML changes: empty fields, garbage data, or duplicates after a site update
• Blocking: IP bans, 403s, bot challenges (proxy + rate-limit handling — I don't bypass paywalls or secure auth)
• Flaky runs: Selenium/Playwright timeouts, hangs, memory leaks
• Broken APIs: endpoint/pagination changes, bad JSON parsing
• Data integrity: encoding issues, missing fields, broken CSV/JSON/DB exports

My QA background means I don't just patch the bug — I find the edge case that broke it and add error handling so it won't break the same way again.

Proof: I built and maintain uaculturehub.com — a live data product that aggregates dozens of sources into PostgreSQL, serves an API, and powers a Next.js UI, managing 40,000+ deduplicated records on automated nightly runs.

Send me your broken script, the target URL, and the error logs — let's get your data flowing again.
Data Tool
Python
What's included
Service Tiers Starter
$45
Standard
$120
Advanced
$250
Delivery Time 2 days 3 days 4 days
Number of Revisions
000
Andrii V.Status: Offline

About Andrii

Andrii V.Status: Offline
AI, Data & QA Engineer | RAG, Pipelines, Test Automation
Kyiv, Ukraine - 3:57 pm local time
AI APPS, AGENTS & RAG - I build the AI itself, not just the data for it: assistants that answer from your own documents, task automations, and tool-using agents. Applying my 13 years of software QA obsession to AI - every answer is grounded in a cited source or honestly says "not found", and I include a golden-set eval harness that checks accuracy, grounding, and prompt-injection resistance on every run, so you can test it yourself. AI that is evaluated, not hoped.

I rescue your scrapers when they break - and build resilient data pipelines that don't fail silently.

13 years in software QA taught me one thing: data is only useful if it is reliable. That obsession drives the jobs clients hire me for:

SCRAPER RESCUE - Scraper dead, returning garbage, blocked, or broken after a site change? I diagnose the real cause (layout shift, anti-bot, rate limits, parser bug), fix it, and validate the output with real test cases - same-day diagnosis, typical fixes in 24-48h. You stop losing data today.

DATA PIPELINES, BUILT TO LAST - Custom scrapers and ETL that clean, deduplicate, validate and load your data into PostgreSQL, Google Sheets, or your own DB, on a schedule. QA-gated from day one: no duplicates, no empty required fields, and failures that alert you instead of silently corrupting data.

Why my data is different: I never hand over a raw dump. Every dataset is validated - deduplicated, format-checked, source-verified, and clearly flagged where data is not publicly available.

AI / RAG-READY DATA - Building a RAG app or fine-tuning a model? I deliver clean, structured, QA-validated data ready to ingest: deduplicated, normalized, chunked, schema-consistent - no hallucination-inducing junk. 13 years in QA means the dataset is validated before it ever hits your pipeline.

Proof (in my portfolio): I built and run a live data product - a cultural-events aggregator that scrapes dozens of sources into PostgreSQL, serves an API, and powers a fast Next.js front end with search, filters and maps. 40,000+ deduplicated records, automated nightly. The full stack - scrape, clean, validate, store, API, UI - in production.

Toolkit: Python (Scrapy, Playwright, BeautifulSoup, Selenium), PostgreSQL, REST APIs, Pandas, Google Places API, LLM enrichment (Claude / OpenAI / Gemini).

I am building my reputation on Upwork, so my focus is 100% on client success. I deliver QA-verified data and I don't make promises I can't keep. Want proof before you commit? Tell me your broken scraper or your target data, and I will send a free diagnosis or a small sample first.

- Andrii

Steps for completing your project

After purchasing the project, send requirements so Andrii can start the project.

Delivery time starts when Andrii receives requirements from you.

Andrii works on your project following the steps below.

Revisions may occur after the delivery date.

Diagnose

I reproduce the failure and pinpoint the real root cause: selector drift, anti-bot, timing, API change, or parser bug.

Fix & harden

I repair it, replace fragile waits, harden the login and error handling, then re-run end-to-end against your real data.

Review the work, release payment, and leave feedback to Andrii.