Roles & Responsibilities
What You’ll Do :
Project scope / overview :
The intern will design, develop, and deploy AI-powered workflows that automate critical internal processes at Airwallex. The intern will build production-ready agents, integrate them with existing Airwallex tools (e.g., Databricks, Salesforce), and deploy them for live internal use by selected pilot teams.
Detailed Outline :
Technical tasks / deliverables :
- Technical design scoping : Create workflow diagrams, identify dependencies, expected inputs and outputs, and define success criteria (e.g., latency, accuracy, cost).
- Agent development : Build and integrate AI agents (e.g., finance flux analysis agent) with multi-agent orchestration as needed, complying with infosec guardrails
- Evaluation & observability : Develop evaluation framework and set up observability dashboards with logging, metrics, and alerts to monitor agent performance and reliability.
- Deployment : Deploy working agents that have passed success criteria, in production environments for pilot
Project Milestones :
Technical design sign-off : Finalize workflow diagrams, success criteria, and technical design documentation.MVP Agent development : Build and integrate AI workflow agents with initial test runs on sandbox data.Evaluation and observability setup : Develop evaluation framework and set up observability dashboards with logging, metrics, and alerts to monitor agent performance,MVP deployment and showcase : Deploy working agents for pilot team access, and showcase to AI team and VP of Data and AI.Expected learning :
Prompt engineering patterns e.g., structured outputs, few-shot, tool-callingEvaluation best practices e.g., using LLM-as-a-Judge approaches for automated evalsLangFuse for observability, to track key metrics and debug issuesReact framework for frontend developmentLangChain / LangGraph for multi-agent orchestration and workflow graphsModel Context Protocol (MCP) for standardising tools for agentsPlanned role & responsibilities
Actively participate in team standups, sprint reviews, and retrospectives to align with engineering priorities.Collaborate with cross-functional partners (e.g., Operations, Finance, Infosec) to understand workflow requirements and validate outputs.Contribute to code reviews, testing, and documentationDeliver bi-weekly project updates and participate in mid-point and final showcases to demo working agents.Design, develop, and deploy assigned AI agents end-to-end under mentor guidance.Share learnings and recommendations for scaling agentic workflows beyond MVPOutline of onboarding process steps in team :
Buddy assignment : Pair with a senior engineer / mentor for end-to-end feature lifecycle walkthrough, and weekly technical check-ins.Tools orientation : Setup access to relevant AI and development toolsDomain familiarization : Walkthrough of identified finance operational workflows (e.g., cash reconciliation, PO approvals) with relevant domain expertsProposed schedule of Review Check-ins :
Monthly check-in for progress review and feedback sharingFinal Review : at 24 weeks (End-of-project showcase)Eligibility requirements
Able to commit to a 6 month placement period, ideally commence in December 2025 through to June 2026.Able to commit to Full Time hours for the duration of the placementAre a current student in Singapore, studying towards a formal tertiary qualification at a recognised university.Strong verbal and written communication skillsDesire to learn and grow with Airwallex as a technologistWe’re looking for people who meet the minimum skills and experience for this role. The preferred skills are great to have, but are not mandatory.
Minimum skills & experience :
Proficiency in Python, with strong software engineering fundamentals.Ability to design and implement AI-driven systems that automate workflows and improve customer experiences.Strong skills in system design and critical thinking, building elegant abstractions over complex patterns, especially in financial services.Proven track record of writing clean, high-quality, production-grade code and ensuring robust interactions between services.Experience working with production AI systems, adhering to best practices in engineering, security, and AI governance.Excellent communication and collaboration skills, with the ability to partner across technical and non-technical teams.Preferred skills & experience :
Proficiency in Kotlin, Typescript or JavascriptFamiliarity with MLOps practices and deployment pipelines for AI-driven products (e.g., model serving, monitoring, retraining)Experience with vector databases, orchestration frameworks (LangChain, Ray, etc.), or event-driven systems (Kafka, RabbitMQ)Exposure to cloud AI / ML services (AWS, GCP, or Azure), including GPU orchestration and scalable model deploymentApplication and Assessment process
Applications are reviewed on a rolling basis with application status changes and outcomes communicated as soon as decisions are finalised. Your patience during this process is highly appreciated.
Application submission including your current resume and academic transcript / s.Online technical assessment.Talent screenTechnical InterviewHiring Manager interviewRequirements : Please provide us with the PDF copies of your transcripts and CV outlining when you finish your studies when applying
About You :
You can think on your feet – with knowledge of building platforms which resolve fast paced, ever changing success measures.You are a team player– you enjoy working in a collaborative, innovative and customer first environment.You are naturally curious– always asking why to better understand our customer needs.You are a passionate problem solver – You love digging deep and brainstorming with colleagues in order to come up with the best solution for a problemEqual opportunity
Airwallex is proud to be an equal opportunity employer. We value diversity and anyone seeking employment at Airwallex is considered based on merit, qualifications, competence and talent. We don’t regard color, religion, race, national origin, sexual orientation, ancestry, citizenship, sex, marital or family status, disability, gender, or any other legally protected status when making our hiring decisions. If you have a disability or special need that requires accommodation, please let us know.
Airwallex does not accept unsolicited resumes from search firms / recruiters. Airwallex will not pay any fees to search firms / recruiters if a candidate is submitted by a search firm / recruiter unless an agreement has been entered into with respect to specific open position(s). Search firms / recruiters submitting resumes to Airwallex on an unsolicited basis shall be deemed to accept this condition, regardless of any other provision to the contrary. If you are approached by someone claiming to represent Airwallex, please verify with our team.
Tell employers what skills you have
Machine Learning
Azure
Pipelines
AWS
Critical Thinking
Kotlin
TypeScript
Software Engineering
JavaScript
Reliability
Protocol
Logging
System Design
Finish
Apache Kafka
GCP
Orchestration
Technical Design