Fense is an AI-powered platform delivering smart recommendations for real estate and agriculture....Fense is an AI-powered platform delivering smart recommendations for real estate and agriculture....
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Fense is an AI-powered platform delivering smart recommendations for real estate and agriculture. By analyzing data and trends, it helps users make informed investment decisions, optimize yields, and identify high-value opportunities. Predictive insights, risk assessment, and actionable advice make Fense AI a practical solution for smarter, data-driven choices.
Post image
Post image
Post image
Post image
DC's avatar
This looks really solid. The onboarding, map views, and recommendation flows make the product feel quite complete. I also like the earthy green palette, it fits the real estate and agriculture context nicely. 🌱
Taoheed 's avatar
Appreciate that! 🙌, We put a lot of thought into making the onboarding and recommendation flows feel intuitive while keeping the product grounded in the real-world needs of real estate and agriculture. Glad the earthy palette resonated too 🌱
DC's avatar
Yeah, that focus really comes through. Curious how the recommendation system evolves as more user data comes in. 🌱
Taoheed 's avatar
Absolutely. The recommendation engine is designed to get smarter as more user preferences, property data, location signals, and interaction patterns come in. Over time, that should allow FENSE to move from broad recommendations toward much more personalized land and real-estate...
Saad's avatar
Predictive/recommendation models like this usually live behind an API rather than on-device, curious how you're handling latency for the map view refreshes as new investment data streams in. Flutter's isolate model helps keep those recommendation recalculations off the UI thread if that's ever a bottleneck.
Taoheed 's avatar
Yeah, for FENSE, the heavier recommendation logic lives server-side so we’re not tying the map experience to on-device model inference. The client mainly handles visualization and incremental updates as new data comes in. We’re also keeping the data pipeline event-driven so we...
Saad's avatar
Event-driven over polling makes sense there, keeps the map from re-rendering on every tick. Are you pushing those incremental updates over WebSockets/SSE, or is it more of a poll-and-diff pattern? That choice usually decides how much debouncing you need on the map redraw side.
Saad's avatar
Worth watching either way: once updates land as individual events, the map redraw cost usually comes from marker diffing rather than the transport itself. Batching events into a 200 to 300ms window and diffing by id before re-render tends to buy more than swapping transport does.
Taoheed 's avatar
Yeah, that’s a good point. For FENSE, we’re treating the stream and the rendering layer separately, so batching/diffing is definitely something I’d prioritize before changing the transport layer. A short debounce window with ID-based diffs should keep the map responsive while...
Saad's avatar
Debounce plus ID-based diffing is the right fix, keeps render cost tied to what changed not event volume. Viewport culling stacks well on top of that, skips diff work entirely for markers outside visible bounds. Scales clean even if transport swaps later.
Back to feed
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started