Freelance projects

Freelance projects

Automatic parser in Google Sheets - which works by itself, without the involvement of a computer, autonomously.

Data Parsing — incorrectly specified categories?

Project translated automatically. Log in or register, to view the original

TECHNICAL ASSIGNMENT - please send real prices and real deadlines considering that electricity may be turned off

Autonomous lead parsing system for construction job vacancies (PL / EU)

⸻

1. GOAL

Create an autonomous server system 24/7 that collects construction job vacancies from specified sources on a schedule, extracts international phone numbers of employers (with country code), strictly filters vacancies by construction professions, and automatically maintains Google Sheets in a fixed format with uniqueness by phone number.

⸻

2. SOURCES (STRICTLY)

• olx.pl (Praca)

• praca.pl

• pl.indeed.com

• jobs.pl

• aplikuj.pl

• gowork.pl (Praca)

• jobo.pl

⸻

3. AUTONOMY AND OPERATION MODE

• Deployment on a server/VPS/cloud.

• Operation without manual launch and without administrator PC.

• Scheduler for launches (at least once a day; support for N hours/N days).

• Auto-restart in case of failures.

• Logs of launches and errors are mandatory.

⸻

4. REQUIREMENT FOR RESISTANCE TO ANTI-PARSING PROTECTION (MANDATORY)

4.1 Context

Sources apply access restrictions, dynamic loading, hiding contacts, activity limits, and behavioral checks.

4.2 Requirement for result

The system must consistently extract phone numbers when they are available to the user in the website interface and correctly degrade under temporary restrictions (pause/reschedule tasks), without losing data integrity.

4.3 Mandatory implementation properties

• Support for dynamic page rendering if necessary.

• Correct operation with website sessions.

• Dosing of actions affecting access to contacts.

• Separation of stages:

1. collecting the list of vacancies;

2. controlled collection of contacts.

• Adaptive response to temporary restrictions (without "storming").

The technical assignment specifies the requirement for stability and results; specific methods are chosen by the performer.

⸻

5. KEY DATA RULE

1 row in the table = 1 unique phone.

Phone numbers never repeat.

If the phone is absent or does not match the format - the record is not created or updated.

⸻

6. REQUIREMENTS FOR PHONES

• International numbers with country code are allowed.

• Storage format: + and digits (E.164), 7–15 digits.

• Validation: ^\+\\d{7,15}$.

• If the number is provided without a country code - the record is not created.

• Uniqueness is determined by the normalized phone.

⸻

7. GOOGLE SHEETS - FORMAT (STRICTLY)

7.1 Columns and order

1. Name Surname of the representative

2. Client's phone

3. Who is needed

4. E-mail

5. NIP

6. Country

7.2 General rules

• No value → empty cell.

• Default country: Polska.

• Any record/update - only with a valid phone.

⸻

8. UPSERT LOGIC (BY PHONE)

8.1 Algorithm

1. extract phone;

2. if there is no valid international number → STOP;

3. normalize;

4. find row by phone;

• no row → create;

• there is a row → update according to the rules below.

8.2 Field "Who is needed" - aggregation

Element format:

<job_title> / <normalized profession>

Storage: list, each position on a new line in one cell.

Rule:

• if the element is not yet present - add;

• if present - do not add.

8.3 Merge rules for other fields

• Name Surname: fill only if empty.

• E-mail: fill only if empty (or add via ; - according to config).

• NIP: fill only if empty.

• Country: if it was Polska by default and another value is found - update; otherwise do not overwrite.

⸻

9. "BUILDERS" FILTER (STRICTLY)

A vacancy is allowed if:

• at least one trigger word of a construction profession is found in the title or description;

• stop words are absent.

9.1 Profession triggers

The full list of forms provided by the client (PL/RU/UA) is used, without abbreviations.

Search is case-insensitive, by substring.

The list is stored in the config and is extensible.

9.2 Stop words (minimum)

rekruter, hr, biuro, księgowa, handlowiec, sprzedawca, call center, obsługa klienta, marketing, magazynier, produkcja, pakowacz, sortowanie, kurier, taxi.

(Extensible via config.)

⸻

10. COUNTRY DETERMINATION

Search by title/description/location.

In case of a match - normalized value:

Dima, [10/02/2026 02:04]

Polska, Niemcy, Czechy, Słowacja, Węgry, Austria, Belgia, Holandia/Niderlandy, Hiszpania, Francja, Włochy, Szwecja, Szwajcaria, Norwegia, Dania, Finlandia, Irlandia, Wielka Brytania/Anglia (all forms).

If not found → Polska.

⸻

11. ARCHITECTURE

• Server deployment (Docker or systemd).

• Task scheduler.

• Separation of collection stages.

• Storing secrets outside the code.

• Log rotation.

⸻

12. LOGS AND MONITORING

Mandatory:

• launch log (source, time, status);

• error log;

• counters: found vacancies, discarded, new phones, updated rows;

• recording of temporary access restriction events.

⸻

13. CONFIGURATION (WITHOUT HARDCODING)

The config includes:

• sources and schedules;

• dictionaries of triggers/stop words/countries;

• aggregation and merge rules;

• activity limits;

• Google Sheets parameters.

⸻

14. ACCEPTANCE CRITERIA

The system is accepted if:

• operates autonomously on schedule;

• the table is updated regularly;

• phone numbers are not duplicated;

• without a phone, data is not written;

• "Who is needed" is aggregated correctly;

• the filter only passes construction;

• under temporary access restrictions, the system correctly degrades and continues working later;

• logs are maintained.

Proposals 12

Ihor K.

6 0

Budget: 1000 PLN Deadline: 1 day

I have experience parsing tens of thousands of prices from dozens of marketplaces every day, feel free to contact me.

Skepra C

1 0

Projects -
Rating -
Rating 457

Budget: 2000 PLN Deadline: 6 days

Hello, Andrey.

I am ready to develop a standalone server system for parsing construction job vacancies from specified sources. The project will include the creation of API endpoints for automating processes, which will ensure stable operation without manual intervention.

As a solution, I will use Python with the Playwright library for dynamic page rendering, which will allow bypassing anti-scraping restrictions. The system will be deployed on a VPS with the ability for auto-restart and logging of all operations. The scheduler will ensure regular data collection at least once a day.

I have experience in creating similar systems. For example, I developed a parser for collecting data from several websites, which ensured correct extraction of information even in the presence of protective mechanisms. In another project, I set up an automated system that collected data and recorded it in Google Sheets, ensuring the uniqueness of entries.

The project timeline will be 5-6 days. I am ready to discuss the details and terms of cooperation.

Oleksandr Kuzomka

0 0

Projects -
Rating -
Rating 128

Budget: 500 PLN Deadline: 4 days

Hello! I have extensive experience in developing autonomous parsers and automation systems in Python.

📌 What I offer for your project:
• A fully autonomous parser that will operate 24/7 on a server.
• Integration with Google Sheets API for instant data updates.
• Implementation of complex phone validation logic (E.164) and deduplication.
• Flexible configuration through a separate file (sources, schedules, limits).
• Detailed logging of processes for monitoring functionality.
• Setup of automatic execution (cron/custom intervals).

I am ready to implement a system that will be reliable and scalable. Feel free to reach out!

Yevhen Naumenko

0 0

Projects -
Rating -
Rating 264

Budget: 7000 PLN Deadline: 15 days

Good day
the cost for the general system + 1 source 500 euros/5 days
then add 1 source 200 euros/2 days
huge experience, wrote parsers for most of the specified sites, with sending by bot in TG, saving in DB
PHP CURL

Mykhailo P.

123 0

Budget: 15000 PLN Deadline: 60 days

real deadlines and cost: one and a half - two months. $3-4 thousand. if this is acceptable, we can discuss all the details. experience in parsing is available.

Yurii Yurii

2 0

Projects -
Rating -
Rating 196

Budget: 10000 PLN Deadline: 20 days

Hello. I have extensive experience in web scraping. I have created scraping systems with microservice architecture and deployment in cloud environments (AWS). A detailed analysis of the websites is also required, as they may have protection against scraping. I am ready to discuss the task in more detail.

Oleksii Pavlyuchenko

4 0

Budget: 10000 PLN Deadline: 30 days

No matter what anyone says, if I respond without AI from myself, the project is quite large, much larger than it may seem, and it cannot be tackled on a whim.
First of all, an analysis of the sources and where the numbers are located is needed; if they are in the HTML of the website itself, it simplifies and reduces the cost significantly, but I highly doubt they will do it that way. Most likely, they are somewhere in the HTML and somewhere deeper, which makes the project even more labor-intensive.

We will have to assemble a separate mini-factory of programs to make it circulate.
I won't scare you or dissuade you from the idea; it is fundamentally very interesting and has the potential to bring savings if your company has sufficient scale.

The price needs to be discussed.

Joseph Collins Mahihu

0 0

Projects -
Rating -
Rating 421

Budget: 3850 PLN Deadline: 30 days

Hello Dymytro
Based on my understanding, you need a stable, fully autonomous 24/7 server system for parsing construction job vacancies from 7 Polish websites, extracting international phone numbers, filtering strictly by profession, and updating Google Sheets — all while handling anti-parsing measures, power outages, and operating without a local PC.
I have vast experience in creating autonomous systems, and I am confident, that I'm the best fit for the job.
Before, we get to the technical details, kindly answer the following questions:

-VPS: Do you have a preferred VPS provider, or should I recommend one (Hetzner/OVH)?

-Proxies: Will you require rotating residential proxies for high-volume parsing?

-Google Sheets Access: Can you create a service account for the Sheets API and share the template sheet?

-Alerting: Would you like Telegram/Discord alerts for system errors or daily summaries?

-Backup: Should parsed data be temporarily stored in a database before Sheets sync for recovery?

Kindly inbox me, with the answers to these questions, first, then we proceed forward. Looking forward to hearing from you

Nick Osipov

41 4

Budget: 1000 PLN Deadline: 3 days

Good day!

I professionally develop standalone parsers in Python. I am ready to implement your system for Google Sheets taking into account all requirements for stability, anti-scraping, and deployment on the server. I have experience working with Google API, bypassing protections, and server infrastructure.

Write to me in private messages, and we will clarify the details.

Dmytro Puzii
Warszawa, Poland

Projects 4
Contests 1
Rating -
Rating 326