How auriny/ai-together works
omsk-together processes municipal complaint spreadsheets through an asynchronous Java-to-LLM workflow: upload and batch incidents, classify and filter them, aggregate by district and topic, generate district and global summaries, then return a ZIP of Excel and Word reports. The README also describes anonymization and an optional Gradio interface. The sampled excerpts show summarization prompts instructing removal of personal names, but do not establish a separate anonymization stage. The UI and controller behavior are only partly evidenced, so links beyond documented capabilities are omitted.
Main entry points: main.py Batch workers · ExcelService.java Spreadsheet processing · redis.py LLM Redis client · qwen.py Summary pipeline
Components 21
Intake and export
- Incident API
core/src/main/java/ru/auriny/core/controller/IncidentController.java - Spreadsheet processing
core/src/main/java/ru/auriny/core/service/ExcelService.java - Incident records
core/src/main/java/ru/auriny/core/dto/IncidentRow.java - Report rows
core/src/main/java/ru/auriny/core/dto/FinalReportRow.java - Report response
core/src/main/java/ru/auriny/core/dto/ReportResult.java
Queue handoff
- Core Redis client
core/src/main/java/ru/auriny/core/service/AiQueueService.java - Redis queues
- Task batch
llm/dto/batch.py - LLM Redis client
llm/queue_client/redis.py - Model and Redis settings
llm/settings.py
Classification
- Batch workers
llm/main.py - Classification pipeline
llm/pipeline/classifier/rubert.py - RuBERT inference
llm/models/classifier/rubert.py - Classification labels
llm/enums/labels.py
Analysis and summaries
- District aggregation
llm/dto/temp_storages.py - Summary pipeline
llm/pipeline/processor/qwen.py - Qwen generation
llm/models/processor/qwen.py - Summary records
llm/dto/summary.py
User interface
- Gradio web interface
llm/web_ui.py - Report download
Other components
- Municipal analyst
Connections 27
- Municipal analyst → Incident API: uploads workbook
README.md - Incident API → Spreadsheet processing: delegates processing
core/src/main/java/ru/auriny/core/controller/IncidentController.java - Spreadsheet processing → Incident records: creates records
core/src/main/java/ru/auriny/core/service/ExcelService.java - Spreadsheet processing → Core Redis client: pushes batches
core/src/main/java/ru/auriny/core/service/ExcelService.java - Core Redis client → Redis queues: writes and reads
core/src/main/java/ru/auriny/core/service/AiQueueService.java - LLM Redis client → Redis queues: pops and pushes
llm/queue_client/redis.py - LLM Redis client → Task batch: validates task
llm/queue_client/redis.py - Batch workers → LLM Redis client: receives tasks
llm/main.py - Batch workers → Classification pipeline: classifies batch
llm/main.py - Classification pipeline → RuBERT inference: filters text
llm/pipeline/classifier/rubert.py - Classification pipeline → Classification labels: filters labels
llm/pipeline/classifier/rubert.py - Batch workers → District aggregation: accumulates incidents
llm/main.py - Batch workers → Summary pipeline: summarizes districts
llm/main.py - Summary pipeline → Qwen generation: generates summaries
llm/pipeline/processor/qwen.py - Summary pipeline → District aggregation: reads topic statistics
llm/pipeline/processor/qwen.py - Summary pipeline → Summary records: creates summary records
llm/pipeline/processor/qwen.py - Batch workers → LLM Redis client: publishes results
llm/main.py - Spreadsheet processing → Report rows: renders report rows
core/src/main/java/ru/auriny/core/service/ExcelService.java - Spreadsheet processing → Report response: returns ZIP result
core/src/main/java/ru/auriny/core/service/ExcelService.java - Core Redis client → Report rows: decodes queued results
core/src/main/java/ru/auriny/core/service/ExcelService.java - Incident API → Report response: returns response
core/src/main/java/ru/auriny/core/controller/IncidentController.java - Municipal analyst → Gradio web interface: uses interface
README.md - Gradio web interface → Incident API: submits analysis
- Municipal analyst → Report download: downloads archive
README.md - Model and Redis settings → RuBERT inference: configures model
llm/models/classifier/rubert.py - Model and Redis settings → Qwen generation: configures model
llm/models/processor/qwen.py - Model and Redis settings → LLM Redis client: configures Redis
llm/queue_client/redis.py