Data & Engineering
QC Reconciliation Web App
A browser-based QC Reconciliation tool that automatically compares two CSV or XLSX files and reconciles numbers at every granularity level - grand total, single dimension, and multi-dimension…
The challenge
Why it exists
Data and reporting teams spend hours manually reconciling numbers between two systems, downloading files, writing VLOOKUPs in Excel, and checking values row by row. This process is slow, error-prone, and doesn't scale across multiple dimensions or granularity levels.
The approach
How it works
A browser-based QC Reconciliation tool that automatically compares two CSV or XLSX files and reconciles numbers at every granularity level - grand total, single dimension, and multi-dimension combinations. The tool handles real-world complexity including schema differences between files, non-additive metrics like Reach, date format mismatches, and missing rows. Results are surfaced in a structured 5-step wizard with a full mismatch table, granularity tree, row count checks, and drill-down to source rows. Built entirely with Claude - architecture design, code generation, and debugging were all done via Claude Code with no manual coding. The app runs 100% in the browser with no backend or API keys required.
Key capabilities
What it does
A browser-based QC Reconciliation tool that automatically compares two CSV or XLSX files and reconciles numbers at every granularity level - grand total, single dimension, and multi-dimension combinations.
The tool handles real-world complexity including schema differences between files, non-additive metrics like Reach, date format mismatches, and missing rows.
Results are surfaced in a structured 5-step wizard with a full mismatch table, granularity tree, row count checks, and drill-down to source rows.
Built entirely with Claude - architecture design, code generation, and debugging were all done via Claude Code with no manual coding.
Business impact
reduces that to under a minute per run - an estimated saving of 100+ analyst hours per month across a team
Manual reconciliation across reporting systems can typically takes hours per analyst per week. This tool reduces that to under a minute per run - an estimated saving of 100+ analyst hours per month across a team. Additionally, it eliminates human error in multi-dimensional comparisons that are practically impossible to do reliably in Excel.
Built with
Technology
Tools & Frameworks
More in Data & Engineering
Related applications
Data & Engineering
Database Field Mapping & Discovery Tool
Connects to databases Extracts database metadata (schemas, tables, columns, relationships) Generates data dictionaries and ER diagrams Integrates with the Gemini API to enable conversational…
ViewData & Engineering
LLM API Gateway with Observability
Built a set of dedicated AWS Lambda functions, one per LLM provider (Gemini and OpenAI), each acting as a clean API wrapper around that provider's SDK. Any frontend or backend can freely switch…
ViewData & Engineering
SQL Standardization & Optimization Bot
To solve all of the above issue I have created SQL Standard Optimization AI bot. This helps to maintain same standards accross whole team.
ViewWant something like this for your team?
We'll map your workflow and scope a working prototype — typically in three weeks, not three months.
Talk to us