MedLedger
Target User: Patients looking for hassle-free medical appointments, as well as doctors, practice managers, and clinic administrators handling high daily patient volumes. The Problem: Patients face long queues, unpredictable waiting times, and confusion when visiting doctors without clear scheduling. Simultaneously, clinics and healthcare providers suffer financial loss and waste valuable staff hours trying to manually organize patient flow and resolve scheduling conflicts.
Preliminary score
- Value for the user14.0 / 25
A specific problem and a clear outcome from solving it.
- Prototype and use of AI8.0 / 30
A working core scenario, and what AI actually contributes to the result.
- Quality testing7.5 / 20
Tests, examples of failures, and comparison with the current approach.
- Feasibility7.5 / 15
Data requirements, running costs and a clear next step.
- Originality3.0 / 10
A meaningful difference from common solutions.
What the judges wrote
- GPT-6 AstraPreliminary41/100
- Strengths
- Your proposal connects patient waiting-time problems with concrete clinic burdens: manual scheduling, conflicts, and wasted staff hours. Your testing write-up identifies specific failures, including vague symptom handling, hallucination-prone inputs, and stale debt records. Your planned 3–5-clinic Baku pilot, tiered pricing, and hosting budget provide a concrete starting point for adoption.
- To improve
- Your materials need one consistent core scenario: the deck emphasizes scheduling and analytics, while the testing describes billing and diagnostic assistance. Document an end-to-end run with sample inputs, saved results, and the AI contribution clearly distinguished from ordinary scheduling logic. Provide results from the 50 simulated scenarios, including failed outputs and evidence that the fallback works, alongside a comparison against current clinic workflows. Specify required clinic data, patient-data safeguards, and per-clinic hosting and AI costs, with a measurable pilot target such as reduced staff handling time.
- Claude Opus 5.5Preliminary39/100
- Strengths
- Your problem framing is relatable: clinic queues, double-booking and manual shift management are real pains for private clinics in Baku. Your business model is concrete, with Solo, Clinic and Pro tiers, a 9,006 AZN investment breakdown including Hetzner hosting, and a sensible roadmap to pilot with 3–5 clinics. Your testing notes honestly name specific failures, such as vague symptom inputs, hallucinations on non-standard terms with a doctor-fallback fix, and offline debt sync not updating, which is better than a generic success claim.
- To improve
- Your submission describes two different products: the deck shows a scheduling and patient-flow platform, while your testing covers an AI diagnostic assistant, hardware-ID licensing and debt recording. Show one coherent core workflow end to end, with screenshots or a demo walkthrough in the deck. AI appears only as one 'analytics' bullet in the deck, so state exactly what the model does, which model it is ("GPT 6 Astra" is not a recognisable model), and why the result depends on it. Report your 50-scenario test as numbers (how many correct, how many fell back) and compare against current clinic practice, for example minutes of wait time or registrar hours saved per day.
The team
Ayşən HəmidovaLead
data
Zəhra Rüstəmova
developer
Davud İsayev
business
Elshan Hamidov
product