Scrapers that keep running months after delivery

Full-time Python developer specialising in large-scale web scraping, Apify actor development and AI-powered automation. I do not just write a script and hand it over: I build scrapers that survive anti-bot systems, resume after crashes, and keep running months after delivery.

Selected client work

Problems clients brought me, what I built, and what changed once it was running. Each card opens the full write-up.

01

Large-Scale E-Commerce Price Tracker

Built a Scrapy + Playwright pipeline scraping 500K+ product listings daily across 20 sites, with change detection and PostgreSQL storage.

ScrapyPlaywrightPostgreSQLCelery
02

AI Document Processing Agent

LangChain-powered agent that extracts, classifies and summarizes documents from multiple formats with GPT-4 integration.

LangChainOpenAIPythonAI
03

B2B Lead Generation Engine

Automated LinkedIn + directory scraping to build verified CEO/CFO lead lists with email validation achieving 95%+ deliverability.

Lead GenPythonEmail Verification
04

AutoTrader MCP Apify Actor for LLMs

Built a custom Apify actor exposing live AutoTrader car listings as an MCP tool, letting GPT, Claude, and Gemini query real vehicle data in natural language.

ApifyMCPLLMAutoTraderGPT
05

Telegram Notification Bot

Feature-rich bot with real-time alerts, data dashboards, user management, and REST API integration for 10K+ subscribers.

Telegram APIPythonREST API
06

n8n Business Automation Suite

End-to-end workflow automation connecting CRM, email, Slack, and custom APIs - eliminating 20+ hours of manual work per week.

n8nAPI IntegrationAutomation
07

Product Comparison Tool

Full-stack comparison engine scraping multiple retailers, normalizing specs, and generating automated Excel/PDF reports for clients.

PythonExcelData Analysis

Jobs where the site fights back

Anti-bot protection, pages that change after they load, results that paginate for ever. Three pieces from the portfolio on my Freelancer profile.

Anti-bot scraper: 1M+ records

A high-volume extraction job against a site that actively blocks automated traffic. The scraper collected more than one million records and delivered them as spreadsheet-ready files.

  • Rotating proxies, so no single IP carries the load
  • Randomised request headers and human-like pacing
  • Output delivered as Excel and CSV
anti-bot-scrape.jsondelivered
records
1,000,000+
challenge
anti-bot protection
approach
rotating proxies, randomised headers, human-like pacing
output
Excel, CSV

Marketplace and directory scrapers on Apify

A suite of Apify actors for listing sites and directories. Each one walks every results page, turns what it finds into one structured dataset, and runs in the cloud with nothing for the user to install.

  • Pagination followed to the last page of results
  • One consistent schema per source, ready to export
  • Published on the Apify Store, so anyone can run them
Run the vehicle actors
marketplace-actors.jsonlive
platform
Apify Store
sources
AutoTrader, Kijiji, AutoScout24, PistonHeads, Canada411, Fnac
paging
automatic
output
structured Apify dataset

Browser automation for real user journeys

An automated test suite that walks through the flows a customer takes: logging in, browsing, adding to the cart and checking out. It handles elements that appear or change after the page loads, which is where brittle scripts break.

  • Login, browsing, cart and checkout covered end to end
  • Dynamic page elements handled reliably
  • Repeatable runs that catch a broken flow before customers do
user-flows.jsondelivered
type
automated test suite
flows
login, browse, cart, checkout
elements
dynamic, handled explicitly

Work you can run right now

Alongside client projects I publish my own scrapers on the Apify Store. They are public, priced per result, and I ship the fix when a target site changes. These numbers refresh from the Store every six hours.

public actors
48
total runs
150,188
users
2,211
average rating
5.0
Browse all 48 actors

Most used, by distinct users

  1. Bulk AI Image Generator (No API Key) It generates AI images in bulk from a list of text prompts - no API key needed. Ideal for producing large batches of… 398 users
  2. Bulk LLM Runner - GPT, Claude, Gemini, Perplexity (No API Key) It runs your prompts in bulk across GPT, Claude, Gemini and Perplexity - no API keys required. Upload a list of promp… 367 users
  3. Autotrader Canada Scraper Extract complete car listings from Canada's largest auto marketplace 296 users
  4. Spitogatos.gr Real Estate Scraper Scrape 25+ attributes per Greek property listing in English & Greek 296 users
  5. Canada411.ca Business Contact Scraper It crawls Canada411.ca and extracts business and people contact information from every listing - names, phone numbers… 170 users
  6. Wallapop Marketplace Scraper (ES/IT/PT) It scrapes Wallapop second-hand marketplace listings at scale across Spain, Italy and Portugal. Paste any search or i… 130 users

Six things I get hired for

Most projects combine two or three of these: a scraper feeding an AI step, or a lead list delivered through a bot.

  1. Apify actor development and publishingInput schemas, pricing and Store listings, built for people who will never read the code.
  2. Data extraction and advanced web scrapingPlaywright, Selenium, Scrapy or BeautifulSoup, picked per site rather than by habit.
  3. AI-powered applications and agentsLLM steps that classify, extract and summarise what a scraper brings back.
  4. Custom automation workflowsPython and n8n pipelines that take over a job someone was doing by hand.
  5. B2B lead generation and list buildingSourcing from directories and company sites, then enriching and deduplicating.
  6. Telegram bots and API integrationAlerts, reports and controls delivered where the team already works.

The stack behind these projects

PythonPlaywrightSeleniumScrapyBeautifulSoupLangChainOpenAIn8nTelegram Bot APIFastAPIFlaskPostgreSQLMongoDBDockerGitPandasNumPyRequestsaiohttpCelery

172 reviews, 5.0 average

A handful, copied from my Freelancer profile, where every review comes from a completed project.

β˜…β˜…β˜…β˜…β˜…

Exceptional work! Youssef not only fixed the existing Python script but also updated it to improve performance and accuracy. He worked efficiently, communicated clearly, and delivered the project ahead of schedule. His attention to detail and problem-solving skills were very impressive. Highly recommend him for any Python development or data scraping projects!

Fix and update python script
Rahul K.Pune, India
β˜…β˜…β˜…β˜…β˜…

Very high quality work. Very responsive and adapted changes very quickly. Understood requirements very well. Very quick work. Very impressed and would recommend.

Product Comparison Tool
David G.Bangor, United Kingdom
β˜…β˜…β˜…β˜…β˜…

He completed the task as required and on time. Excellent job!

URL Email Search Task
William C.Sisters, United States
β˜…β˜…β˜…β˜…β˜…

Fast and optimal work.

Python Web Scraper for Publisher
Christoph S.Feldafing, Germany
β˜…β˜…β˜…β˜…β˜…

Good communication and response time. Highly recommend!

Get email leads of bus companies in Sweden
Jens H.FΓ€rjestaden, Sweden

From a URL to data you can rely on

  1. 01 / scope Tell me the site and the fields

    Send the URL, the data you need and how often you need it. I reply within 24 hours with what is feasible and a quote.

  2. 02 / build Built against the real site

    Retries, logging and structured output are in the first version, not bolted on after the first failure.

  3. 03 / harden Made to survive

    Anti-bot handling where the site needs it, and crash recovery so an interrupted run resumes instead of starting over.

  4. 04 / support Handed over, not abandoned

    You get the code or the running actor with a walkthrough, and I stay reachable when the target site changes.

Have a site that needs scraping?

Describe the job and I will tell you honestly whether it is feasible, usually within 24 hours. If one of my public actors already does it, I will point you there instead.