Senior Python AI Backend / Runtime Engineer

MTicket is building a production AI runtime on top of a real ticketing and live events business.
This is not a chatbot wrapper, not a demo, and not a pet project.
Our systems operate around real-world workflows: ticket sales, payments, stadium shows, scanners, live event operations, approvals, and time-sensitive business processes.
We are looking for a strong Senior Python Backend Engineer who can build reliable AI-powered infrastructure in production.

 

What you’ll build

You’ll work on the core AI runtime layer, including:
 

• Async AI workers and orchestration runtime
• Event-driven pipelines and queue processing
• LLM integrations with OpenAI / Anthropic
• LLM proxy layer and structured output validation
• Retries, timeouts, DLQ and failure handling
• RAG pipelines, pgvector and vector search
• Approval flows and human-in-the-loop workflows

• Observability, tracing and production debugging

   

Tech stack

Python, FastAPI, LangGraph / LangChain or similar orchestration runtimes, OpenAI / Anthropic, PostgreSQL, pgvector, Kafka / SQS / RabbitMQ / Redis, OpenTelemetry, AWS, Docker.

 

Must have

• Senior-level production Python experience
• Strong FastAPI and async Python skills
• Experience with event-driven systems: queues, workers, retries, DLQ
• Solid PostgreSQL experience
• API integrations in production
• Production debugging experience
• Understanding of observability, tracing and monitoring
• Ability to own complex backend systems end-to-end

• Ukrainian or Russian language proficiency at C1 level.

   

Strong plus

• LLM pipelines in production
• LangGraph, LangChain or other orchestration runtimes
• RAG, embeddings, pgvector or vector search
• OpenTelemetry
• Structured outputs and JSON schema validation
• Kafka, SQS or RabbitMQ

• Experience with payments, ticketing, marketplaces, live events or operational workflows

   

First 90 days

In your first 90 days, you will:
 

• Build async worker infrastructure
• Implement the orchestrator runtime
• Integrate the LLM proxy layer
• Build validation pipelines
• Connect telemetry, tracing and observability

• Ship the first operational AI workflows into production

   

What you’ll get

• Architectural ownership from day one
• Real production AI systems, not demos
• Direct impact on AI runtime architecture
• Strong engineering culture
• Fast decision-making and close collaboration with leadership

• Remote work from anywhere 

   

This role is a great fit for an engineer who wants to build real AI infrastructure: queues, workers, orchestration, observability, validation, production reliability and systems that directly affect live business operations.