Tap items on the right to ring up sale.
Live Kitchen Display System (KDS)
Real-time live queue for Shastri Nagar barista station
๐ Operational Analytics & Pattern Discovery
Recognizing what is being sold, when it is being sold, rush heatmaps & customer basket affinities
๐ง Algorithmic Pattern Discoveries & Operational Takeaways
โฐ When It Is Being Sold: Operational Dayparts & Rush Shifts
Breakdown of demand by shift: morning beverages, lunch comfort, evening snack surge, and late-night social dining.
๐ Hourly Sales Trajectory (08:00 - 23:00)
Hourly transaction distribution highlighting peak customer arrival windows
๐ Day-of-Week Rhythm
Revenue velocity comparison across Monday through Sunday
๐ What Is Being Sold: Category Performance
Revenue contributions, items ordered, and peak consumption hours per category
| Category | Revenue | Share | Top Selling Hero | Peak Hour |
|---|
โก High-Velocity Products
Fastest selling menu items, estimated gross margins, and dominant ordering windows
| Menu Item | Units | Revenue | Margin | Selling Window |
|---|
๐ Customer Basket Affinities (Frequently Bought Together)
Natural item pairings appearing in the same order โ ideal for cashier cross-sell prompts and combos.
๐ง AI Temporal Intelligence & BI Engine AWS Database Integrated
Fine-tuned database AI model evaluating cafe operations, demand baselines, and BI tools for any given time
High-throughput, sub-10ms transactional engine. Manages counter billing, inventory decrements, and cashier authentication.
Serverless storage replication. Automatically syncs commits from Aurora to Redshift in <10s without ETL pipelines or POS lag.
Analytics brain. Powers Apache Iceberg Time Travel queries, Redshift ML demand predictions, and SageMaker anomaly detection.
Apache Iceberg on S3 & Redshift Time-Travel Console
FOR SYSTEM_TIME AS OFQuery "how data looked at any particular point in time" with native Iceberg snapshot metadata without impacting Aurora OLTP performance.
s3://hourglass-pos-analytics-866874944450/iceberg/
| Snapshot Columns |
|---|
| Click "Execute Time Travel Query" or choose a preset to reconstruct historical data. |
Fine-Tuned AI: Redshift ML & SageMaker Operational Engine
CREATE MODEL SQL + SageMaker AnomalyMachine learning models trained directly inside Redshift SQL. Integrates with SageMaker to detect real-time sales anomalies, forecast peak demand, and prescribe tactical combos.
cafe_hourly_demand_model_v2
Evaluating real-time customer stream...
AI Operational State: Evaluating...
--
--
Tomorrow's Forecast: -- Orders
Peak rush window: --
โ Morning Shift Kitchen Prep List:
๐ BCG Menu Engineering Matrix
๐ Market Basket Analysis (Apriori Association)
| Primary Item | Associated Item | Confidence | Lift Multiplier |
|---|
๐ฅ RFM Customer Segmentation Matrix
| Customer | Phone | Segment | Recency | Visits | Spend | Action Recommendation |
|---|
๐ Sales Ledger & Historical Reports Full Audit Trail
Query, filter, and audit past transactions (Active Cashier: Samay Kashyap)
| Ticket # | Date & Time (IST) | Cashier | Type | Customer | Items Ordered | Subtotal | Discount | Total (โน) | Payment | Action |
|---|
Gemini & Antigravity AI Business Strategist Active Live Telemetry
Algorithmic intelligence & business decisions grounded in Hourglass Cafe's PostgreSQL sales database
Quarterly Financial Health
90-day gross revenue stands at โน2,20,658 across 1,259 tickets (AOV: โน175). Evening 4 PMโ8:30 PM window produces 58% of daily turnover.
Stars vs Dogs Re-Engineering
Stars: Adraki Chai, Classic Cold Coffee, Vegilicious Sandwich.
Dogs: Sweet Corn Chat, Mint Sandwich.
RFM Churn Mitigation
Identified 8 high-value patrons approaching churn risk. Milestone 10th-visit reward ready for active regulars.
Ingredient Requirements & Runways
| Ingredient | Category | Current Stock | Yesterday (T-1) | Day Before (T-2) | Today Left (T) | Tomorrow (T+1) | Day After (T+2) | Runway | Status | Rec. Restock | Action |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Loading predictive forecast... | |||||||||||