CareerVault

What it is

CareerVault turns resumes, notes, accomplishments, and user interactions into versioned canonical profiles. Those profiles support grounded question answering, job matching, tailored application documents, and portable exports.

Request CareerVault access or sign in if you already have an account. Access requests are reviewed manually while billing remains disabled.

Why it exists

The aim is to reduce the repeated work of reconstructing a career history for every application while keeping generated material grounded in the user’s own record.

My role

I designed and built the product and its career-archive workflow end to end through Praxish, my independent applied AI and data science practice.

Technical shape

The application uses FastAPI and PostgreSQL with pgvector, SQLAlchemy and Alembic migrations, and database-backed asynchronous workers. Its workflows include retries and deduplication, workspace role-based access control, artifact tracking, and usage and cost telemetry.

CareerVault ingests source material, compares that record with job descriptions, asks follow-up questions, prepares tailored resume and application material, and keeps the underlying record portable to other tools.

Decisions and constraints

Application material should remain grounded in the source archive rather than inventing experience. The canonical profile provides the reference point for grounded question answering, job matching, and tailored documents.