Banner Image

Skills

  • Artificial Intelligence
  • Data Extraction
  • Data Management
  • JSON
  • Programming
  • Python
  • API
  • Automation Engineering
  • FastAPI
  • Large Language Models
  • Microsoft Azure
  • Reports
  • Research
  • Research Papers
  • Retrieval-augmented Generation

Sign up or Log in to see more.

Services

  • Evidence-Grounded PDF Data Extraction to

    $25/hr Starting at $25 Ongoing

    Dedicated Resource

    I provide AI-assisted, evidence-grounded data extraction from complex PDF documents, especially annual reports, sustainability/ESG reports, financial reports, and other corporate documents. You provide...

    APIArtificial IntelligenceAutomation EngineeringData ExtractionData Management

About

Python & AI Developer | RAG, PDF Data Extraction, FastAPI & Automation

I am a Python and AI developer focused on practical automation, document intelligence, data extraction, and LLM-powered applications.
I build solutions that turn complex documents and repetitive data-processing tasks into structured, reliable workflows.
One of my main projects is an evidence-grounded document analysis system designed for complex annual and sustainability reports. The system processes long PDFs, retrieves relevant evidence for predefined questions, generates structured answers, preserves source-page references, and returns NOT_FOUND when the available evidence is insufficient instead of guessing.
My work includes:
• Python automation and scripting
• PDF and document data extraction
• Structured Excel, CSV, and JSON outputs
• Retrieval-Augmented Generation (RAG)
• LLM integrations and structured outputs
• FastAPI and REST API development
• Vector search and document retrieval
• Data processing and transformation
• Retrieval and evidence-quality improvements
• Debugging and improving existing Python applications
I can work on both custom software development projects and managed document-analysis projects where you provide the documents and required fields/questions and I deliver structured, source-backed results.
I focus on clear requirements, maintainable implementations, reliable outputs, and avoiding unnecessary complexity.

Work Terms

I prefer clearly defined project scopes, deliverables, and acceptance criteria before work begins.
For document-analysis and data-extraction projects, clients should provide the documents, required questions/fields, preferred output format, and any domain-specific rules or terminology. For larger batches, I recommend starting with a small paid pilot before processing the full dataset.
Deliverables may include Excel, CSV, JSON, structured reports, source-page references, evidence, scripts, APIs, or other agreed outputs depending on the project.
For managed document-extraction services, my pre-existing RAG/document-processing engine, internal tools, prompts, retrieval logic, and source code are not included unless explicitly agreed in advance.
For custom software development projects, source-code delivery will be defined separately as part of the project scope.
Revisions cover corrections and adjustments within the originally agreed scope. New requirements, additional documents, new fields/questions, or significant scope changes may require a new milestone or revised quote.
I communicate primarily through written updates and provide clear progress, questions, and blockers during the project.