Browse categories
Explore
Fiverr Pro
English
$
USD
I will do sql database migration using claude code ai agents
Kelson
About this gig
I build AI-agent-driven data pipelines for database migration, cleanup, and validation using Claude Code, the same approach I use professionally on production systems. I've used this pipeline to migrate 11.4M+ records across 5 business modules into a unified PostgreSQL schema, spanning all 14 company branches, with no existing ERDs to start from and no records lost or duplicated across branches, cutting migration time from 8 weeks to under 2 weeks per module. I'm currently extending the same pipeline to a sixth module, branch by branch.
For your project, I'll use the same discipline: schema mapping, four-layer automated validation (structural, volumetric, semantic, and referential integrity), a documented policy for resolving any data conflicts, and a clear audit trail, so you can see exactly what changed and why.
Good fit if you need: messy or multi-source data consolidated into one clean database, a migration between database engines (MySQL, SQL Server, PostgreSQL, SQLite), or an automated cleanup and validation pass on an existing database.
Get to know Kelson
Kelson
Data Engineer
- FromIndonesia
- Member sinceMay 2023
Languages
English, Indonesian
I'm Kelson, a Computer Science student and Data Engineer building AI-agent-driven pipelines for database migration and cleanup. Using Claude Code, I've architected a pipeline that handles schema inspection, lineage mapping, script generation, and multi-layer validation end to end, migrating 11.4M+ records across 14 branches with zero data loss. I bring the same discipline to freelance work: clear scoping, a documented audit trail, and validation you can actually check, not just a promise. If your data is messy, undocumented, or scattered across sources, I can help you consolidate it properly.
FAQ
What information do you need before we start?
Read access to your source database(s), plus any existing documentation. If there's no ERD, I can reverse-engineer the structure directly from the live data, that's part of the process.
Is my data kept confidential?
Yes. I only access what the migration needs, never share or reuse your data.
Can you work on a live production database?
Yes, though I recommend running the migration against a staging copy or during an off-peak window so your live system isn't affected during the process.
What if my project doesn't fit these packages?
Message me with your specifics. Scope, tables, and pricing are easy to adjust for larger, multi-source, or more complex migrations than what's listed.
Which database engines do you support?
PostgreSQL, MySQL, SQL Server, and SQLite, including migrations between any combination of these.
What's included in the validation report?
A structural, volumetric, semantic, and referential integrity check confirming nothing was dropped, duplicated, or mismatched during migration.
