Survey Analysis Platform
Turn mixed survey data into quantitative insight and qualitative highlights.
- Role
- ML / backend engineer
- Year
- 2021
- Timeline
- Research + delivery
This is an NDA-signed client project, so it isn’t linked publicly. I’ll walk you through it live on a call and answer anything about how it was built.
A research team had rich survey responses across text, numbers, images and video, but no fast way to move from raw responses to insight and shareable highlights.
The data was genuinely multi-modal, and 'the interesting bits' had to be found semantically, not by keyword.
Used semantic search to surface and reel together highlights.
Instead of Keyword filters over transcripts.
Why: Semantic search found relevant moments even when the words didn't match, which is where qualitative insight actually lives.
Stored the mixed responses in a document model instead of rigid tables.
Instead of Forcing text, numbers, images and video into one relational schema.
Why: Survey data is genuinely irregular; a document store let each modality keep its own shape without a schema migration every time a question changed.
Split quantitative and qualitative analysis into separate passes.
Instead of One monolithic analysis step over everything.
Why: Numbers want aggregation and free text wants NLP; separating the passes let each use the right tool and kept every result explainable back to its source.
- Multi-modal ingestion (text, numeric, image, video)
- Quantitative + qualitative analysis
- Semantic-search highlight & showreel generation
- Dockerised deploy on EC2 with Nginx
Hemant Manglani
Ahmedabad, India
Have a product that has to work? Let’s talk.
A 20-minute call, no pitch. We work out whether this is a real problem worth solving, and if I’m not the right person, I’ll tell you.
Every message answered within one business day