Senior Software Engineer, AI Agents

This is a fully remote position, open to applicants in Singapore.

📋 Description

• Transition our card operations agent from internal testing to full-scale production — establishing the reliability, observability, and safeguards necessary for a system managing real financial data and personally identifiable information (PII).

• Manage the trade-offs between latency, model selection, cost, and safety — making practical architectural choices that maintain our unit economics as we expand.

• Develop agent systems that are proactive, rather than reactive — crafting solutions that foresee the needs of finance teams instead of waiting for requests.

• Enhance agent functionalities across accounting automation, policy enablement, and card operations — collaborating with the AI Product Lead to prioritize initiatives that significantly boost adoption.

• Remain informed on the latest advancements in LLM research and tools — assess new models, methods, and architectures, integrating effective solutions into our technology stack.

• Approach work as a product engineer, not solely as an AI engineer — every system you create should facilitate platform adoption and enhance clients' experiences in measurable ways.

⛳️ Requirements

• Over 8 years of experience in full-stack or backend development — with a strong emphasis on Python as your primary programming language.

• Proficiency in Java, Golang, or Rust is also appreciated.

• Solid foundation in software engineering — experience in designing distributed systems, APIs, and scalable backend architectures.

• 1–2+ years of practical experience in building AI agents for production — you have progressed beyond merely prompting LLMs and have delivered systems that can reason, plan, and take action.

• Comprehensive understanding of LLM internals — you know how models function beneath the surface, not just how to interface with an API.

• Architectural insight beyond just frameworks — you can determine when tools like LangChain, LangGraph, AutoGen, or other agent frameworks are appropriate, and when to build from foundational principles.

• Knowledge of Model Context Protocol (MCP) — you recognize what MCPs are and how they facilitate agent-tool interoperability.

• Leadership and delegation abilities — you are adept at guiding and reviewing the work of others, taking ownership of outcomes rather than merely tasks.

• Strong communication skills — you can convey complex technical decisions in a clear manner to both engineering and product stakeholders.

🏝️ Benefits

• Insurance coverage following probation.

• Reap Card stipend.

• Access to AI tools in the workplace, along with opportunities to learn, experiment, and grow with them.

• A culture centered on innovation, inclusion, and continuous learning.