Agentic AI Engineer

Join AiMi and build the future of Capital Markets Infrastructure

At AiMi, we are revolutionising the management of critical trading and market data infrastructure through AI-driven automation. By leveraging advanced technologies, we replace manual, static processes with dynamic, data-driven workflows that enable organisations to achieve greater efficiency, agility, and control.

We operate at the cutting edge of agentic AI - deploying autonomous agents that ingest venue and vendor notifications, assess impact across client infrastructure, and drive change management workflows in real time. If you’re technically strong, AI-curious, and want to ship meaningful work on a live production platform, we want you on the team!

Tasks

What you’ll do:

As an AI Software Engineer, you will work across the full stack of the AiMi platform. Building and shipping features from intuitive front-end interfaces through to robust back-end services and the agentic AI workflows that sit at the heart of our product. You will own features end-to-end, contribute directly to our agent pipeline, and play an active role in shaping how we build software as a team.

Key responsibilities include:

• Build and ship new features and functionality that directly serve our clients: across back-end services, agentic workflows, and integrations that power the platform.
• Contribute to the development and continuous improvement of our agentic AI pipeline, translating complex capital markets workflows into reliable, production-grade agent behaviour
• Design, build, and maintain skills, connectors, and MCP-based integrations that extend the capabilities of AiMi agents — connecting to client systems, internal tooling, and third-party platforms
• Contribute to agent evaluation - designing and running evals that assess agent accuracy, reliability, and behaviour across real-world workflows, and using those insights to drive continuous improvement
• Build and maintain production observability and alerting across the agent pipeline, including monitors for agent latency, error rates, failed workflows, and unusual failure patterns, integrated with team paging/notification workflows.
• Build resilient agent and streaming workflows, including appropriate timeouts, failure handling, retries, circuit breakers, fail-fast behaviour, and recovery patterns for long-running LLM, agent, and SSE-based calls.
• Diagnosing dependency and version conflicts in Node.js/npm monorepos, including package resolution, transitive dependencies, lockfiles, and runtime.
• Participate in agentic SDLC practices: spec-driven development as part of day-to-day engineering workflow
• Build and maintain robust back-end services container and serverless based services and APIs in Node.js , with a working understanding of our React/TypeScript front-end codebase
• Work across the full stack: React/TypeScript front-end, Node.js back-end services, and AWS Lambda - to deliver well-tested, maintainable code and scalable services.

Requirements

What we’re looking for

• 3+ years of hands-on development experience
• Solid back-end development experience in Node.js, with experience building and consuming RESTful APIs
• Practical experience working with LLMs, agent frameworks, or AI automation tooling: Mastra, LangChain, LlamaIndex, Claude, OpenAI or similar
• Familiarity with agentic SDLC practices and spec-driven development workflows (Claude Code, Cursor, CodeRabbit, or similar)
• Working knowledge of AWS cloud services - Lambda, Bedrock, DynamoDB, S3, Cognito, API Gateway, CloudWatch
• Experience with MongoDB and DynamoDB (or equivalent NoSQL databases)
• Solid understanding of event-driven architecture patterns
• Comfortable with testing at multiple levels - unit, integration, and E2E - and a genuine commitment to developer led testing approaches
• Familiarity with CI/CD workflows, GitHub Actions, and version control best practices
• A pragmatic, ownership-driven mindset with the ability to work independently in a remote, fast-paced environment
• Clear written communication skills - important for async collaboration and for working within our spec-driven development process

Good to have

• Domain knowledge in capital markets, trading infrastructure, market data, or financial services
• Experience with LLM observability tooling such as Langfuse, Helicone, or similar
• Familiarity with streaming architectures and server-sent …