The Umanitek Guardian Agent protects people from deepfakes and digital impersonation by combining Decentralized Knowledge Graphs (DKG) with perceptual intelligence. If you’re excited by the idea of building resilient infrastructure and operational systems for agents that reason, not just predict, you’ll feel right at home.

As a Backend Engineer, you'll work on the Guardian Agent server-side architecture which includes the backend services, REST APIs, and plugins that power them. You’ll contribute directly to building the microservice architecture from the ground up, develop MCP-based agent tooling, and transform AI-driven capabilities into reliable, production-ready products.

You’ll spend a significant portion of your time building and extending the REST APIs and MCP tools utilized by the Guardian agent, while also making sure that all the proper security, scalability and maintainability principles are always in place.

This is a hands-on role where ownership, architectural thinking, and the ability to move fluidly between responsibilities really matter.

Responsibilities

• Build and maintain backend features for the Guardian ecosystem, spanning APIs, MCP servers and agent logic
• Develop and extend microservices that expose MCP tools, REST endpoints, and agent capabilities
• Implement backend services using Node.js + Express and Nest.js REST APIs to support Guardian workflows
• Ensure security, performance, observability, and reliability across the stack
• Participate in product discussions and help shape how users and agents interact with AI safety systems
• Maintain alignment between Guardian features and the broader Umanitek roadmap
• Ensure the architecture is fully scalable, secure and compatible with modern cloud computing services

Technical Experience

• 4+ years of professional software development experience in a backend engineering role
• Strong experience with TypeScript across backend services
• Experience building backend services with Node.js, Express and Nest.js
• Experience with LLM systems/applications and the Model Context Protocol (MCP)
• Familiarity with REST APIs, authentication flows, and event-driven or agent-based systems
• Experience working with APIs, automation, scripting, and API/SDK integrations
• Experience with relational databases and ORMs (TypeORM/Prisma/Drizzle)
• Comfort working across repositories and collaborating in interdisciplinary teams
• Strong communication skills and a product-oriented mindset

Nice to Have

• Strong experience with Nest.js framework
• Experience with cloud computing services (AWS/GCP)
• Experience working with Decentralized Knowledge Graphs, semantic web, or graph data
• Familiarity with Web3 / blockchain concepts (wallets, signatures, decentralized infrastructure)
• Experience with automation pipelines or agentic workflows
• Basic familiarity with Turborepo build system

Tech Stack and Tools

• Typescript
• Node.js, Express
• Nest.js Framework
• REST APIs, MCP tools & agents
• GitHub, CI/CD
• Docker
• Turborepo
• SQL Databases & ORMs
• AWS / GCP (nice to have)

Why umanitek:

We are solving problems that are technically hard and structurally unsolved: hallucinations, impersonation, coordinated abuse, and misinformation embedded directly into AI systems, where privacy, jurisdiction, and scale actually matter. You work with people who have already built companies and shipped complex systems, in an environment with high standards, real ownership, and freedom to push ideas into production. You shape the product, the architecture, and the direction of the company from day one. Think American-scale ambition, built in Europe, with the goal of defining how AI safety is actually done.

The position is fully remote