İnolook
We are building Qoshul.az for urban Gen-Z youth (ages 18–30) who struggle with digital isolation and waste hours coordinating offline activities in chaotic messaging groups, as well as local HORECA and sports venues facing 60%–80% empty capacity during off-peak hours. Today, users face high friction, frequent last-minute cancellations, and safety concerns when trying to meet up offline, while venue owners burn marketing budgets on non-targeted ads instead of attracting active, nearby customers in real time.
Preliminary score
- Value for the user16.0 / 25
A specific problem and a clear outcome from solving it.
- Prototype and use of AI13.0 / 30
A working core scenario, and what AI actually contributes to the result.
- Quality testing8.5 / 20
Tests, examples of failures, and comparison with the current approach.
- Feasibility5.0 / 15
Data requirements, running costs and a clear next step.
- Originality5.5 / 10
A meaningful difference from common solutions.
What the judges wrote
- GPT-6 AstraPreliminary52/100
- Strengths
- Your combination of hyper-local activity matching and off-peak venue offers connects a consumer coordination problem to a concrete business need. Your testing write-up identifies relevant failure modes—cold-start timeouts, indoor GPS errors, and margin-eroding discounts—rather than relying only on success claims.
- To improve
- Your next iteration should demonstrate one complete venue workflow, from capacity input and AI group formation to a margin-safe offer and verified check-in, including actual model inputs and outputs. Your testing would be stronger with case counts, recorded failures, and measured coordination time or venue utilization against the current approach; the stated baselines do not establish an improvement. Your feasibility plan should include per-event operating costs, consent and retention rules for location/profile data, and a narrowly scoped pilot with one venue.
- Claude Opus 5.5Preliminary44/100
- Strengths
- Your problem framing is two-sided and concrete: Gen-Z users who waste time coordinating meetups in chaotic chat groups, and HORECA and sports venues with empty off-peak capacity. Pairing social matchmaking with venue yield management gives the product a clear business model. Your testing write-up names realistic failure modes: cold-start matching in low-density areas, indoor GPS errors triggering false anti-abuse blocks, and discount recommendations hurting margins during unexpected rushes. Those are the kinds of problems a real deployment would hit.
- To improve
- Your materials do not show what the AI actually computes. Explain how the Match Score is produced (Gemini prompt? embeddings? rules?) and show real inputs and outputs, so judges can tell AI logic from Lovable-generated UI. Back your testing and comparison with numbers: how many test users or events, match success rate, moderation precision, and measured coordination time versus the claimed 2+ hours in WhatsApp. Add running costs (Gemini calls, Supabase, Stripe fees), a data plan for the cold-start problem, and one concrete next step such as a pilot with named Baku venues. Also name the venue-side metric (e.g., off-peak occupancy uplift) that fits the Enterprise track.
The team
Sadiq Suleyman Islamov
developer
Amin NuriyevLead
product