Mr Golden

TELEGRAM CHATBOT WITH N8N AUTOMATION

Dec 2025 - Present • 8 mos

This project addresses a major challenge: managing high-volume customer inquiries without sacrificing personalization or data integrity. To solve this, I engineered an end-to-end chat automation ecosystem that turns a Telegram bot into an intelligent, data-driven assistant. By integrating Telegram, n8n, OpenAI, Supabase, and PostgreSQL, the system autonomously handles complex conversations and structures the resulting data in real time. The workflow activates the moment a user messages the Telegram bot. An n8n trigger webhook captures the inbound payload, extracting the text, sender details, and timestamp. The engine routes this data to an OpenAI chat model node. Using advanced prompting, the AI analyzes user intent, crafts a highly contextual, human-like response, and instantly streams it back to the user via the Telegram API, maintaining a fluid conversation 24/7. Simultaneously, the backend handles robust data persistence. The n8n workflow parses the conversation log, separating user inputs, AI responses, and extracted profile attributes. It then initiates secure database connections to store this structured data. User accounts and authentication states are synchronized within Supabase, while full chat histories and interaction metrics are written directly to a PostgreSQL database. This architecture eliminates manual data entry and response delays, ensuring every touchpoint is handled professionally while providing a structured repository of user data ready for analytics. #n8n #WorkflowAutomation #AIAutomation #LowCode #NoCode #APIIntegration #Webhooks #BusinessAutomation #DataIntegration #BackendDevelopment #DevOps #OpenSourceAutomation #AIChatbot #OpenAI Integration #Supabase #PostgreSQL #MakeAlternative #ZapierAlternative #SoftwareEngineering #AutomationEngineer #ProcessAutomation #CloudAutomation #DigitalTransformation #SaasAutomation #EnterpriseAutomation #DataPipelines #Serverless #CustomIntegrations #TechStack #WorkflowOptimization

DevOps Infrastructure Alerting & Incident Response Matrix

Jun 2026 - Jul 2026 • 1 mo

Minimizing system downtime depends entirely on how quickly an engineering team can triage infrastructure alerts. To optimize incident response, I architected a DevOps alerting matrix that aggregates server logs, filters out systemic noise, and escalates critical bugs to the on-call engineering squad. By integrating Prometheus, n8n, Slack, and PagerDuty, the setup ensures that infrastructure anomalies are addressed before impacting end users. The workflow monitors health metrics constantly, activating when a Prometheus alert rule fires an HTTP POST request to n8n. The automation engine parses the alert payload, checking the severity level and infrastructure tags against a lookup table. Minor errors are neatly logged to a designated Slack channel for review. However, if a critical database or server outage is detected, the workflow bypasses standard logs, opens a high-priority incident inside PagerDuty to trigger on-call phone alerts, and spins up a dedicated incident response Slack channel populated with the relevant error logs. This automated triage loop removes the lag time between system failures and developer awareness, drastically reducing Mean Time to Resolution (MTTR) while keeping communication channels perfectly organized. #n8n #DevOps #IncidentResponse #PagerDuty #SlackAPI #Prometheus #InfrastructureAutomation #SysOps #WorkflowAutomation #LowCode #APIIntegration #Webhooks #ProcessAutomation #CloudAutomation

MAKE COM AUTOMATON WITH GHL FORM SUBMISSION

Sep 2025 - Jul 2026 • 10 mos

This project addresses a critical vulnerability in modern sales pipelines: delayed lead engagement. When a high-intent prospect submits an inquiry, their interest drops rapidly with every minute of silence. To solve this setback, I engineered an enterprise-grade voice automation ecosystem that instantly bridges the gap between digital marketing capture and direct sales outreach. By integrating GoHighLevel, Make.com, and Bland AI, I built a system that transforms static form submissions into live, intelligent telephone conversations in under sixty seconds. The workflow triggers the exact moment a prospect completes a form or survey within a GoHighLevel funnel. A native automation immediately dispatches the lead payload via a custom webhook to Make.com, which serves as the central orchestration engine. Make.com processes the raw inbound JSON data, parsing vital fields such as the lead's name, phone number, and specific business needs. The system then initiates a secure API call to Bland AI, dynamically injecting these personal details into a specialized, conversational system prompt. Within moments of submitting the web form, the lead receives an automated phone call from a natural, human-like Bland AI voice agent. Operating twenty-four hours a day, the AI agent greets the prospect by name, references the exact details from their form submission, answers immediate service or pricing questions, and executes qualification logic to vet their intent. This setup eliminates manual dialing delays, ensuring every single lead receives an instant, professional, and consistent brand experience while freeing human sales teams to focus entirely on closing pre-qualified opportunities.