Freelance projects

Freelance projects

Python script for bulk uploading documents from two APIs

Databases & SQL, Python — incorrectly specified categories?

218 USD

Client's review of cooperation with Illia Antipiev

Python script for bulk uploading documents from two APIs

218 USD

Ilya, thank you for your work. The project was not easy, but you managed. It took much more time than planned. Completing the task at 100% required patience and many corrections as well as changes from our side. In any case, thank you!

4.5

Maximilian D

Reciprocal Feedback

30 November 2025

Freelancer's review of cooperation with Maximilian D

Python script for bulk uploading documents from two APIs

218 USD

Thank you for the collaboration! All materials and accesses were provided on time and updated as necessary. The client was accommodating when they had to postpone execution for personal reasons. A small downside - sometimes I have to answer questions from a person who lacks context. So I had to repeat myself 😔 but credit should be given - they find some mistakes.

4.6

Illia Antipiev

Reciprocal Feedback

1 December 2025

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

For the mobile application to work, a database is required, which the LLM will use after the user's request in the chat. The database must be in Supabase. If there is no answer in the database, then the LLM will refer to an external source via API.

1. Project Description

Develop an asynchronous Python script for bulk uploading documents:

SITE API to upload all available documents of all types for the entire period.
SITE API: upload all documents starting from the year 2000.

Store the full text, all fields, and metadata of each document. The target database is Supabase (PostgreSQL).

2. Functional Requirements

Tasks for the performer on Python 3.8+:

A. Asynchronous Bulk Upload via API

Use aiohttp + asyncio for asynchronous operation.
Parallelization of requests (10–50 simultaneous connections, without exceeding the allowable API limits).
Support for pagination (pageSize maximum for SAOS — 100, for Sejm clarify in the documentation).
For SAOS implement filtering judgmentDate >= 2000-01-01.
Save all received fields — meta and full text of the decision.

B. Saving in Supabase

Batch saving of data (batch insert, up to 1000–5000 records at a time).
Use supabase-py for integration.
Develop SQL schema:
- Separate tables for sejm_documents and saos_judgments.
- Store metadata in JSONB.
- Indexes for key fields (date, court, document type).

C. Reliability and Process Control

Automatically resume progress after a failure (checkpoint file).
Retries on failures (up to 3 times, with exponential backoff).
Detailed logging — execution time, received objects, errors.

D. Vectorization for Search

After import — chunking each document (25 chunks/document; size to be discussed, approximately 1500–2000 characters).
Store chunks in a separate table (document_chunks), referencing the original document.
(Optional) Form for further vectorization via LLM API (Gemini Flash 2.5 or another equivalent).

3. Input Data

Sejm API: all documents (according to documentation), all types, all years.
SAOS API: all courts, court decisions since 2000.
Target DB: new Supabase project (PostgreSQL), account and keys provided by the client.
Expected quantity: 160,000+ , 520,000+.

4. Data Structure (SQL Schema)

Table: sejm_documents

sql
CREATE TABLE sejm_documents (
  id BIGSERIAL PRIMARY KEY,
  source_id TEXT UNIQUE NOT NULL,
  document_type TEXT,
  title TEXT,
  content TEXT,
  metadata JSONB,
  created_at TIMESTAMP DEFAULT NOW(),
  updated_at TIMESTAMP,
  is_processed BOOLEAN DEFAULT FALSE
);

Table: saos_judgments

sql
CREATE TABLE saos_judgments (
  id BIGSERIAL PRIMARY KEY,
  source_id TEXT UNIQUE NOT NULL,
  court_type TEXT,
  case_number TEXT,
  judgment_date DATE,
  text_content TEXT,
  metadata JSONB,
  created_at TIMESTAMP DEFAULT NOW(),
  updated_at TIMESTAMP,
  is_processed BOOLEAN DEFAULT FALSE
);

Table: document_chunks

sql
CREATE TABLE document_chunks (
  id BIGSERIAL PRIMARY KEY,
  document_id BIGINT REFERENCES sejm_documents(id) ON DELETE CASCADE,
  chunk_index INT,
  chunk_text TEXT,
  created_at TIMESTAMP DEFAULT NOW()
);

5. Deliverables (what must be included in the result)

Asynchronous Python script (+ detailed instructions for running)
SQL scripts for creating the necessary tables
README for deploying the project from scratch
.env.example for configuring Supabase keys
Log file, checkpoint file for monitoring progress
(Optional for MVP): chunking test documents for RAG/LLM indexing verification

6. Requirements

Documentation for the script!
Easy launch for the owner without programming experience.

Proposals 5 Rejected 15

Please select

Jakub Czajka

0 0

Projects -
Rating -
Rating 169

Budget: 8500 PLN Deadline: 20 days

Good morning,

I propose the development of an asynchronous Python script for bulk loading documents from the Parliament and SAOS API, storing them in Supabase, and preparing them for further vectorization for LLM. The script will support batch insert, checkpoints, retries, and detailed progress logging.

I offer a **quote of 8,500 PLN net** with an estimated completion time of **3–4 weeks**.

In my work, I will use asyncio + aiohttp for parallel data fetching, supabase-py for database integration, and I will also develop an SQL schema and a document chunking system, ensuring easy deployment and complete documentation.

I have experience in Python, asynchronous bulk data fetching scripts, and working with PostgreSQL/Supabase, which allows for a stable and scalable implementation of the entire process.

I would be happy to schedule an online meeting to present the implementation plan, the method of chunking documents, and to consult on integration with LLM and API.

Kostyantin Grinyuk

0 0

Projects -
Rating -
Rating 192

Budget: 820 PLN Deadline: 2 days

Hello!
I have experience in Python, asyncio, and Supabase, worked with big data and APIs, ready to perform your project efficiently and quickly.

Roman Z.

7 0

Budget: 1000 PLN Deadline: 1 day

Good day!
My name is Roman, and I am among the top 5 developers in the category of "Artificial Intelligence and Machine Learning" among ~1600 specialists on the platform.
I guarantee:
- Fast and high-quality execution of the task
- Clear adherence to deadlines
- Regular communication throughout the entire process
I would be happy to discuss the details of your project in private messages.

Illia Antipiev

Winning proposal

63 2

Budget: 820 PLN Deadline: 7 days

Hello
I can complete your project
I will write good documentation
For easier deployment, I can create a Docker container

Orest K.

31 0

Projects 32
Rating -
Rating 758

Budget: 850 PLN Deadline: 7 days

Good day!
I have commercial experience working with Python for 3+ years.
I have worked with Supabase and created automation scripts. I am ready to complete your project.

I suggest improvements from my own experience - to use a circuit breaker + retry for API requests. Also, instead of SQL scripts for creating tables, use migrations. Regarding data optimization, I have a few ideas that I would be happy to discuss.

I have only a few questions about the AI part; I do not fully understand what is required.

Michael Yanislav

0 0

Projects -
Rating -
Rating 205

Budget: 820 PLN Deadline: 2 days

Dear Maximilian,

My name is Mikhail, and I am a developer with extensive experience in web application development, automation, and data collection. I would be happy to offer my services for the successful completion of your project. From your specifications, I understand that the best solution for you would be to write a script in Python using the following stack: requests/selenium, sqlalchemy, asyncio/threading. I am a professional in the field of automation and have written many projects related to parallel parsing; it does not matter how complex the resource is from which data needs to be extracted, it will be extracted with maximum speed and quality. To bypass API protection, I will use proxies, and at the end, I will provide a filled database created through sqlalchemy and all the code, and if necessary, I will connect neural networks. I am confident that I can implement your ideas and bring the project to a successful conclusion.

I would be glad to have the opportunity to discuss your project in more detail and answer any questions you may have.

Alona Baryliuk

14 3

Budget: 820 PLN Deadline: 3 days

Hello, I have reviewed your task and I am interested in its implementation, I would like to collaborate with you. I invite you to a personal meeting for a more detailed discussion.

Vadym Shrub

6 0

Budget: 780 PLN Deadline: 1 day

Good day, I am ready to take on your project. I have skills in Python.

Daniils Fjodorovs

2 0

Projects -
Rating -
Rating 475

Budget: 900 PLN Deadline: 6 days

ready to help you out

i think maybe you can use go instead of python

it looks better for this usecase

Nataliia M.

6 3

Budget: 799 PLN Deadline: 11 days

Good day, I will do everything as you say. I hope for cooperation, write in private messages!

Gayane M.

0 0

Projects -
Rating -
Rating 162

Budget: 820 PLN Deadline: 1 day

Hello.
I was interested to learn about your project. I am confident that I can do effective and quality work that meets your requirements and expectations. I have over 8 years of experience. I am ready to discuss the details and start working. I look forward to your response.

Viktor Safronov

1 0

Projects -
Rating -
Rating 292

Budget: 820 PLN Deadline: 2 days

Hello, I have extensive experience in web development. I am ready to do it quickly and efficiently.
Message me privately – we will discuss the details.

Ihor Doronin

9 0

Budget: 1000 PLN Deadline: 3 days

Good evening, Maximilian!
In general, the task is clear, but for an accurate response regarding deadlines and price, I would like to clarify some questions that arose after analyzing your task.
Please write in private messages – we will discuss the details and your wishes.

Dmitro F.

25 0

Budget: 4000 PLN Deadline: 25 days

Hello! I propose to implement it in Go, divide it into sprints, and start with the simple (exclude vectorization, chunks, retries in the early stages), because otherwise, the project may not be completed. Minimum starting price.

Dmytro Zmenkov

1 1

Projects -
Rating -
Rating 121

Budget: 820 PLN Deadline: 1 day

Hello! I am ready to complete this project with extensive experience in developing various applications.

Jeo Vincent C.

4 2

Budget: 820 PLN Deadline: 7 days

Hi,
I'm excited to apply for the role involving asynchronous document ingestion and Supabase integration. With deep experience in Python (3.8+), aiohttp, asyncio, and supabase-py, I can deliver a robust, scalable ETL pipeline tailored to your API and database needs.

I’ve previously built similar systems for high-volume document processing, including pagination, batching (1k–5k inserts), checkpoint-based recovery, and JSONB metadata storage in PostgreSQL. I also understand the importance of chunking and structuring documents for future vector-based search and LLM integration.

You’ll receive a fully-documented, production-ready solution — complete with schema scripts, .env templates, logs, and retry logic — designed for ease of deployment even by non-developers.

Looking forward to contributing to your project.

Best regards,
Jeo Vincent Carretas

Vladislav T.

35 1

Budget: 888 PLN Deadline: 2 days

Hello, I am the one you need.
I have extensive experience in web development.
Message me privately to discuss the work.

Oleksii V.

0 0

Projects -
Rating -
Rating 342

Budget: 820 PLN Deadline: 2 days

Good day.

I am ready to complete your task quickly and efficiently.

Advantages:
- Ease of use.
- Free support for 2 weeks after the order is completed and error corrections.
- I start working on the day the order is accepted and complete it in the shortest possible time.

I will be happy to cooperate.

Oleh Ovchynnykov

52 1

Budget: 820 PLN Deadline: 1 day

Good day

I am ready to do your work
I will be happy to help with your task quickly and efficiently

Maximilian D
Warszawa, Poland

Projects 31
Contests 1
Rating 4.9
Rating 1 102

Maximilian D

Illia Antipiev

Jakub Czajka

Kostyantin Grinyuk

Roman Z.

Illia Antipiev

Orest K.

Proposals are currently absent

Michael Yanislav

Alona Baryliuk

Vadym Shrub

Daniils Fjodorovs

Nataliia M.

Gayane M.

Viktor Safronov

Ihor Doronin

Dmitro F.

Dmytro Zmenkov

Jeo Vincent C.

Vladislav T.

Oleksii V.

Oleh Ovchynnykov

Proposals concealed

Current freelance projects in the category Databases & SQL