A detailed plan for Team - 5 (Sentinels).
| Day | Time Slot | Participant (Role) | Assigned Task | Assignee | Task Details & Goals | Status | Queries |
|---|---|---|---|---|---|---|---|
| Day 1: Foundation & Core Data Pipeline | |||||||
| 1 | Morning | P1 (Frontend) | Screen 1 UI: Upload Component | Naghul | Create Angular component for file upload. Implement drag-and-drop UI, file type validation (.csv), and loading indicators. Build the 'Post-upload Info Summary Card'. | Completed | |
| P2 (Backend) | API Endpoint: /api/data/upload | Naghul | Create .NET endpoint. Receive `IFormFile`, act as orchestrator by forwarding the file to the ML service. Await metadata response and pass it back to the frontend. | Completed | |||
| P3 (ML) | ML Endpoint: /process-csv | Ragavi & Mahalakshmi | Create FastAPI endpoint. Use `pandas` to parse CSV, add `synthetic_timestamp` column. Calculate metadata (records, columns, pass rate, date range). Persist processed data (Parquet). Return metadata JSON. | Completed | What storage method to use? Suggesting Parquet |
||
| P4 (DevOps/Docs) | Project Scaffolding & Docker Setup | All Members | Initialize Git repo. Create monorepo structure. Create initial `Dockerfile` for all 3 services and the `docker-compose.yaml` file. Ensure services can communicate via service names. | Completed | |||
| 1 | Afternoon | P1 (Frontend) | Screen 2 UI: Date Ranges & Visualization | Naghul & Harsha | Build the 'Date Ranges' component with calendar pickers. Implement 'Validate Ranges' button logic. On API success, display summary cards and render `ngx-charts` timeline. | Completed | |
| P2 (Backend) | API Endpoint: /api/data/validate-ranges | Naghul | Create .NET endpoint to proxy date range validation requests from Angular to the Python service. Return the rich validation data to the frontend. | Completed | |||
| P3 (ML) | ML Endpoint: /validate-ranges | Ragavi & Mahalakshmi | Create FastAPI endpoint. Load processed data. Implement validation logic (sequential, no overlap). Calculate record counts for each range and monthly aggregations for the timeline chart. | Completed | |||
| P4 (DevOps/Docs) | Integration Testing & Documentation Start | All Members | Run `docker-compose up --build` and test the full Screen 1 flow. Troubleshoot any networking issues. Start the `README.md` and create the initial System Architecture diagram. | Completed | |||
| Day 2: The Intelligence Engine & Explainable AI (XAI) | |||||||
| 2 | Morning | P1 (Frontend) | Screen 3 UI: Train Screen | Naghul & Harsha | Build the basic UI for Screen 3. Add the 'Train Model' button and implement loading state management (e.g., show spinner on click). | Completed | |
| P2 (Backend) | API Endpoint: /api/model/train | Naghul | Create the pass-through .NET endpoint for model training. Ensure it handles asynchronous, long-running requests gracefully. | Completed | |||
| P3 (ML) | ML Endpoint: /api/model/train (Core) | Ragavi & Mahalakshmi | Implement core training logic. Load data, filter by date ranges. Train XGBoost model. Evaluate performance (Accuracy, Precision, etc.). Save the trained model artifact. | Completed | Which subset of numeric features should be prioritized? | ||
| P4 (DevOps/Docs) | Refine Dockerfiles & Data Flow Diagram | All Members | Optimize Dockerfiles with multi-stage builds. Create the detailed Data Flow (Sequence) diagram for the "Train Model" user journey. | Completed | |||
| 2 | Afternoon | P1 (Frontend) | Screen 3 UI: Final Visualization | Naghul & Harsha | On API success, populate performance metric cards. Render charts. Display the Base64 SHAP plot image received from the API. | Completed | |
| P2 (Backend) | Task: Support XAI Data | Naghul | Ensure the DTOs (Data Transfer Objects) between the backend and frontend can handle the Base64 string for the SHAP plot without issues. | Completed | |||
| P3 (ML) | XAI Integration: SHAP Plot Generation | Ragavi & Mahalakshmi | Enhance the `/train` endpoint. After training, use `shap` to calculate feature importance. Generate a summary plot, save to an in-memory buffer, and encode as a Base64 string in the API response. | Completed | |||
| P4 (DevOps/Docs) | End-to-End Testing (Screen 1-3) | All Members | Perform a full user journey test: Upload -> Configure -> Train. Verify all data and visualizations appear correctly. Document any bugs found. | Completed | |||
| Day 3: Real-Time Simulation & Final Delivery | |||||||
| 3 | Morning | P1 (Frontend) | SignalR Service & Screen 4 UI Setup | Naghul & Harsha | Install `@microsoft/signalr`. Create an Angular service to manage the WebSocket connection. Build the static layout for the Simulation screen (Screen 4). | Completed | |
| P2 (Backend) | SignalR Hub & Simulation Service | Naghul | Implement `SimulationHub` with SignalR. Create the background service with a producer-consumer pattern to read data, call the ML service for predictions, and broadcast results via the hub. | Completed | |||
| P3 (ML) | ML Endpoint: /simulation-ws | Ragavi & Mahalakshmi | Create a low-latency websockt for simulation. Load the model into memory on application startup to avoid reload overhead. | Completed | |||
| P4 (DevOps/Docs) | Finalize Design Document | All Members | Complete the `Design Document.pdf`. Embed the finalized architecture and data flow diagrams. Write concise descriptions for all design choices and the API contract. | Completed | |||
| 3 | Afternoon | P1 (Frontend) | Live Simulation Dashboard | Naghul & Harsha | Connect the Screen 4 component to the SignalR service. As data streams in, update all UI elements in real-time: stats, charts, and the live prediction table. | Completed | |
| P2 (Backend) | Integration & Stress Testing | Naghul | Work with Frontend and ML leads to ensure the entire simulation pipeline is robust. Test for dropped connections and error handling. | Completed | |||
| P3 (ML) | Model & Code Cleanup | Ragavi & Mahalakshmi | Review all Python code for clarity and comments. Ensure the model artifact is saved and loaded correctly. Finalize feature selection logic. | Completed | |||
| P4 (DevOps/Docs) | Finalize README & Demo Video Script | All Members | Write the final, comprehensive `README.md` with foolproof deployment instructions. Write the detailed script and shot list for the 3-minute demo video. | Completed | |||
| 3 | Evening | All Hands on Deck | Demo Video Recording & Final Submission | All Members | Record the screen capture for the demo video (P1). Record the voiceover (P4). Edit the video (P2). Final code review and merge to main (P3). Zip and submit all artifacts. | Completed | |