Hermes: Telegram Operations and Community Workflow Platform
> Private repository. Available for code review on request.
▍ Problem Space
Businesses relying on Telegram for community engagement face fundamental scalability constraints:
- Manual Moderation Bottleneck: Human operators cannot maintain 24/7 engagement across multiple channels.
- Platform Policy and Reputation Risk: Automation around public communities must respect platform limits, reduce spam risk, and keep brand review boundaries explicit.
- Context-Blind Responses: Simple keyword-triggered bots produce generic replies that damage brand perception.
- Workflow Identity Management: Business workflows need state visibility, explicit permissions, and conservative delivery controls.
▍ Architecture
CONTENT WORKFLOW
Channel Monitor → Context Extractor → Draft Generator (Gemini/Claude)
↓
GOVERNANCE LAYER
Platform Rules | Review Queue | Audit Trail | Delivery Policy
↓
MODERATION WORKFLOW
Context Tracking | Approval Boundaries | Quality Review | Incident Notes
↓
TELEGRAM API (MTProto via Grammers)Key Components:
- Channel Monitor: Real-time event stream from target Telegram channels via MTProto.
- Context Extractor: Semantic analysis of post content, thread history, and channel tone.
- Draft Generator: Context-aware draft synthesis with persona-specific voice, tone matching, and review-ready output.
- Governance Layer: Per-workflow publishing limits, review gates, audit trails, and delivery-policy handling.
- Moderation Workflow: Typed workflow state for context tracking, approval boundaries, quality review, and incident notes.
▍ Key Engineering Decisions
Problem
Community operations need several permissioned workflow identities without losing auditability or confusing production ownership.
Solution
External context is parsed into typed records with explicit ownership, review state, and conservative delivery configuration.
Alternative Rejected
Unstructured workflow handling — makes permissions, support, and incident analysis harder.
Problem
Burst publishing creates spam risk, poor audience experience, and avoidable platform friction.
Solution
Conservative delivery governance: workflow-level pacing, review controls for risky actions, and explicit approval before public posting.
▍ Tech Stack
Backend
Rust, Tokio, Grammers (MTProto), Axum
AI
Gemini/Claude via Antigravity Gateway
Data
PostgreSQL (event sourcing)
Governance
Platform rules, review queues, audit trail
▍ Demonstrated Competencies
Platform-Safe Automation
Systems that reduce manual work while preserving review boundaries, rate limits, and brand reputation.
LLM Integration
Production-grade draft generation with context-aware prompting and multi-provider fallback.
Protocol Engineering
Direct MTProto implementation via Grammers for monitoring, moderation context, and internal operational tooling.
Workflow State Management
Typed operational records with review state, approval boundaries, and graceful degradation.
Event-Driven Architecture
Real-time event processing with backpressure and fire-and-forget persistence.