DeepTable – an API that converts messy Excel files into structured data
Agent-guided compilation handles merged cells and multi-level headers LLMs choke on.

Automated CSV cleaning with AI insights, but OpenRefine handles larger datasets for free.
Non-technical professionals, analysts, students working with messy spreadsheets
OpenRefine · Julius AI · Rows
I’m Riki, and I built Recoonlytics after repeatedly running into the same problem: before doing any meaningful analysis, I was spending way too much time cleaning messy datasets.
Missing values, duplicates, inconsistent columns — it often took longer than the actual analysis.
Recoonlytics is an attempt to remove that friction.
You upload a CSV or Excel file, and it automatically:
- detects column types and structure - removes duplicates and fixes missing data - generates quick statistics and summaries - creates simple visualizations instantly
The goal is simple: go from raw data → usable insights in seconds, without manual preprocessing.
One challenge I faced was handling inconsistent real-world data (mixed types, noisy entries). I worked on making the system adaptive so it can clean data without requiring user configuration.
It’s still early, and I’m actively improving it.
I’d really appreciate feedback on:
- how accurate the cleaning is - whether the workflow feels intuitive - what’s missing for your use case
Thanks!
Agent-guided compilation handles merged cells and multi-level headers LLMs choke on.
Survey builder with one-click sentiment analysis, but Qualtrics, Pulse Surveys, and Culture Amp dominate HR tech.
CSV cleaner with deterministic transformations, but remove.bg pattern—solves friction, not novel.
Useful for quick cleanup, but JinaAI and LLMs already handle this natively.
Excel-to-dashboard automation, but Tableau and Power BI already do this better.
Before/after slider motivation for neurodivergent families—no guilt, just progress visualization.