A Community Platform operating in Delhi (NCR region)

01 / The challenge

Start with the real problem.

A small volunteer moderation team was overwhelmed by the volume of flagged posts and comments as the community grew.

02 / The solution

A system built to fit.

I built a toxicity classification model that auto-actions clear-cut cases and routes ambiguous ones to human moderators. Keeping the interface simple enough for Community Platform's non-technical staff to use directly was as much a design constraint as the underlying model itself.

1. 01

   Began by auditing what Community Platform already had in place, thier platform had video , image & text based data that needed moderation, text moderation was kind of easy compared to others

2. 02

   Built and validated an initial version for Community Platform using Python, testing it against real historical data before anything went near production.

3. 03

   Once the core system was working, the focus shifted to integration — making sure it fit into how Community Platform's team already worked, not the other way around.

4. 04

   Closed out the engagement with Community Platform with a handover session covering how the system worked and what to watch for as usage scaled.

03 / The results

What changed.

• Automated moderation on ~70% of flagged content
• Cut moderator backlog significantly during peak activity
• Kept humans in the loop for all ambiguous cases
• Community Platform engaged for a smaller follow-on scope after seeing the initial results