Mira (@mira) is a personal AI agent that acts from your messenger. The agent works inside Telegram across personal and group chats with zero setup, helping users get things done, including scheduling meetings, summarizing discussions, creating content, and coordinating plans, directly within one of the world’s largest messaging platforms.

You'll be joining Mira at a critical growth stage — we're building AI agents that serve millions of B2C users and enterprise clients. We're looking for a Senior Backend Engineer who has shipped production-grade agent infrastructure and knows what it takes to keep it running at scale. You'll own critical parts of the stack: agent memory, LLM integration, orchestration flows, and the reliability layer underneath it all. If you've wrestled with the hard problems of agent systems — state management, non-deterministic testing, latency under load — this role is for you.

Responsibilities:

• Own and evolve core agent infrastructure: memory, orchestration, and LLM integration pipelines
• Build and scale high-load backend systems in TypeScript/Node.js on Kubernetes + AWS
• Design reliable agent flows that handle real-world edge cases at production scale
• Drive technical decisions on architecture, tooling, and agent system design
• Maintain high engineering standards: testing, observability, and clean ownership

Requirements:

• 4+ years of backend engineering experience with a strong Node.js/TypeScript background
• Hands-on experience with AI agents or LLM-based systems in production — or serious pet projects that show you've gone deep
• Proven high-load production background: you've built systems that scale and stay up
• Experience with Kubernetes and AWS at scale
• Strong ownership mindset — you don't wait to be told what's broken