B2 English, based in Argentina. Payment preferably in USDT. Remote only. Quite a combination, right?
I built an AI-assisted job-search system around Freehire and Hirify. Both platforms offer official API/CLI access for automation, which makes them much easier to connect to an agent workflow. Simple Python scripts collect job listings and run the first pass of filtering against strict rules.
Then agents double-check the requirements, match them against my experience, and tailor my CV to each role. They highlight relevant projects and technologies I've actually worked with. Final review and approval stay with me.
โ 20 applications sent;
โ 8 HR screenings passed;
โ 1 video screening completed and submitted.
๐ Technical interviews are still ahead.
I'm not ready to write off the market. Even with my constraints, things are moving.
A job search deserves ๐๐ง ๐๐ง๐ ๐ข๐ง๐๐๐ซ๐ข๐ง๐ ๐๐ฉ๐ฉ๐ซ๐จ๐๐๐ก ๐ญ๐จ๐จ: clear rules, automation, and checks on the results.
AI Assistant Using Your Business Knowledge Base โ RAG on Your Documents
THE PROBLEM
Q&A bots break down when knowledge lives in documents: a 100-page manual has no "questions" to match, it can't fit into a prompt, and generic chatbots hallucinate instead of admitting what they don't know.
THE SOLUTION
A RAG (Retrieval-Augmented Generation) knowledge base: documents are split into meaningful chunks, embedded into a vector index, and the assistant answers from the right sections โ by meaning, not keywords.
Any format as-is: PDF, DOCX, TXT, Markdown โ 100+ pages is fine
Answers grounded in YOUR documents โ it says "I don't have that information" rather than inventing
Source references โ every answer shows which document and section it came from
Runs on your infrastructure โ documents never leave your control
One command to re-index after updating documents โ documented, no programmer needed
The 'I don't have that information' line is the part most RAG builds skip, and it's the one that matters. How do you set the cutoff? On mine, a fixed similarity threshold broke once I filtered results by user role. Scores shifted and it refused questions it could answer.