You will get a tested AI pipeline that turns documents into branded PDF reports
Project details
Most AI work is sold on a promise. I sell it on proof. I build a working pipeline that turns your documents, transcripts, or data into a clean, branded PDF report - and you watch it run on your own sample before you pay.
It stays low-risk by design: it ships in mock mode (no API key, no token cost) with a self-test that must pass, so you verify the whole flow first. The AI prompt lives in an external, editable template, so you change what's extracted without touching code - and swapping in your live OpenAI or Anthropic key is a one-function change. You get clean, commented Python, setup notes, and the full source; nothing is a black box. Hebrew RTL output supported.
I come from algorithmic trading, where one unverified number blows an account - so testing and honest reporting are how I build, not extras. Send a sample input and the format you want, and I'll show it working.
It stays low-risk by design: it ships in mock mode (no API key, no token cost) with a self-test that must pass, so you verify the whole flow first. The AI prompt lives in an external, editable template, so you change what's extracted without touching code - and swapping in your live OpenAI or Anthropic key is a one-function change. You get clean, commented Python, setup notes, and the full source; nothing is a black box. Hebrew RTL output supported.
I come from algorithmic trading, where one unverified number blows an account - so testing and honest reporting are how I build, not extras. Send a sample input and the format you want, and I'll show it working.
AI Algorithms
Large Language Model, Transformer ModelAI Applications
AI Content Creation, Natural Language Generation, Natural Language UnderstandingAI Development Language
PythonAI Tools
Azure OpenAIAI Models
ChatGPT, GPT-4What's included $180
These options are included with the project scope.
$180
- Delivery Time 7 days
- Number of Revisions 1
- AI Model Integration
- Detailed Code Comments
- Natural Language Processing
- Prompt Engineering
- Setup File
- Source Code
About Dror
AI Automation & Integration | Python, OpenAI/Claude | Evidence-Backed
Brissago, Switzerland - 6:07 am local time
I build the AI systems businesses actually need - OpenAI/Anthropic (Claude) integrations, workflow automation, agentic tools with an auditable risk gate, and document/report pipelines (including clean Hebrew RTL PDFs) - in clean, commented code, Python-first and matched to your stack. Then I hand you a tested system and the evidence, so you watch it work before you approve a cent.
What makes my work different is where I come from: algorithmic-trading systems, where a single unverified number can blow up a funded account. That taught me a discipline most automation freelancers skip - backtesting, Monte-Carlo and walk-forward validation, honest reporting that includes what doesn't work. Reliability isn't a feature I bolt on; it's how I build. Every delivery runs clean, ships with a self-test you can run yourself, set-up notes, and sample logs.
How I work: fully written/async, fast turnaround, clear communication. AI-assisted, human-verified - I compile, test, and stand behind every delivery. I'm happy to start with a small free diagnostic or a pay-on-approval first milestone, so you see the quality before you commit.
I'm not the cheapest developer - I'm the lowest-risk one. English and Hebrew.
Steps for completing your project
After purchasing the project, send requirements so Dror can start the project.
Delivery time starts when Dror receives requirements from you.
Dror works on your project following the steps below.
Revisions may occur after the delivery date.
Build, mock-test, then deliver
I confirm your input format and report spec, build the pipeline, and run it in mock mode on your sample so you verify it first. Then I wire in your live key and hand over the source, setup notes, and a short walkthrough.