I’m an AI Engineer focused on building production AI systems end-to-end — from data ingestion and retrieval to user-facing copilots and workflow automation.

I’m currently building an AI-powered meeting intelligence platform across three connected codebases:

• AXIS (Web App)
  A React + TypeScript app where users upload/record meetings, manage transcripts, organize work in spaces, and track AI-generated tasks. app.tryaxisapp.com
• extension_recorder (Chrome Extension, MV3)
  A browser extension for capturing meeting audio from tabs + microphone, authenticating with Supabase OTP, and uploading recordings to the platform.
• supabase_edge_functions (Backend AI + automation)
  Supabase Edge Functions for transcription orchestration (AssemblyAI), task generation (OpenAI), AI chat with session history, and meeting data workflows.

• Designing a full meeting pipeline: audio capture → transcription webhook ingestion → transcript persistence → structured task extraction.
• Building LLM-powered workflows with strict response schemas for reliable meeting-to-task conversion.
• Implementing AI chat over user meeting context with session/message persistence, context snapshots, and token usage tracking.
• Developing multi-surface product architecture (web app + extension + serverless functions) on a shared Supabase backend.
• Strengthening production readiness with auth flows, tenant-scoped data patterns, and secure client/backend boundaries.

Message-driven system that performs retrieval over embedded organizational data to generate LLM-based responses via broker queues.

ETL DAGs using Apache Airflow for data embedding and summarization, with caching and timestamp-based indexing to avoid redundant processing.

Scalable, fault-tolerant workflows using Temporal for async data processing, deduplication, ETL orchestration, and event-driven microservice integration.

Custom LLM fine-tuning and pipeline development for community violation detection and automated reporting.

Multi-agent LLM applications using CrewAI and LangChain for dynamic analysis and orchestrated decision-making.

Python analytics library for behavioral insights, including graph analytics and activity-based segmentation.

• Languages: Python, TypeScript, JavaScript, SQL, LaTeX
• AI/LLM: OpenAI API, llama-index, LangChain, CrewAI, RAG evaluation
• Data & Infra: Supabase, PostgreSQL, Qdrant, Neo4j, MongoDB
• Pipelines & Workflow: Apache Airflow, Temporal, RabbitMQ
• Frontend & Product: React, Vite, Tailwind CSS, Chrome Extensions (MV3)
• Tooling: Docker, Git

• Co-Founder, AI Community Group (Nov 2024 – Present)
  Host a weekly AI series on LLM systems, RAG, agentic workflows, and practical AI engineering.
  🌐 Website

• Co-Founder, Cassandra AI Group (Oct 2021 – Oct 2023)
  Built an academic AI community with workshops, study circles, and a two-day conference.
  🌐 Website | YouTube | GitHub

• ✉️ Email: [email protected]
• 💼 LinkedIn: linkedin.com/in/mramin22
• 💬 Discord: mramin22#1669
• 🐦 Twitter: @mramin22

"After everything, what remains is kindness — so don't hesitate to help others." 😊