AI Engineer

About Thunes

Thunes operates as a cutting-edge global payment network facilitating real-time transactions across 130+ countries and supporting over 80 currencies. The platform connects more than 7 billion mobile wallets and bank accounts worldwide, interfacing with 350+ payment methods including major names like GCash, M-Pesa, AliPay, and WeChat Pay. Its members include prominent gig economy platforms, super-apps, fintech companies, payment service providers, and banks. Thunes prides itself on unmatched speed, control, visibility, security, and cost savings in cross-border payments, enabled by its proprietary Smart Treasury Management Platform and robust compliance infrastructure. Founded and headquartered in Singapore, Thunes also maintains offices globally.

Role Overview

We are seeking a highly motivated AI Engineer eager to contribute to building an advanced "AI Operating System" within a dynamic fintech environment. This position blends startup agility with regulatory discipline, focusing on developing advanced Stateful Agentic Systems and multimodal Retrieval-Augmented Generation (RAG) pipelines capable of interpreting diverse data types including text, charts, and financial tables. The role requires end-to-end responsibility from architecture through deployment of generative AI capabilities linked to our enterprise data lake on AWS. Engineers are expected to manage their deployments independently using modern cloud and container technologies.

Key Responsibilities

• Design and deploy robust RAG workflows using cloud platforms such as AWS and GCP, leveraging frameworks like LangGraph or LlamaIndex for multi-step financial task planning agents.
• Create pipelines for multimodal reasoning that analyze heterogeneous financial documents incorporating charts, graphs, and tables.
• Perform adversarial testing (Red Teaming) to identify vulnerabilities such as Prompt Injection, Jailbreaking, or personal data leaks prior to exposing models to sensitive information.
• Establish and enforce quantitative "Definition of Done" criteria using evaluation tools like DeepEval to assess model faithfulness and relevance, validating model improvements and cost optimization efforts.
• Practice Deployment Autonomy by containerizing applications with Docker, defining infrastructures, and deploying via GitLab CI to cloud environments.
• Implement strict governance guardrails using solutions such as NeMo Guardrails or Guardrails AI to enforce JSON schema compliance and prevent disallowed financial advice.

Required Qualifications and Experience

• A minimum of 5 years in AI engineering, including at least 2 years deploying NLP or Generative AI solutions in production environments.
• Proficient with multi-cloud architecture (AWS, GCP) and mastery of cloud-native LLM ecosystems like AWS Bedrock or Google Vertex AI, with solid knowledge of cloud operations (IAM, Storage).
• Expertise in GenAI orchestration using frameworks such as LangChain, LangGraph, LlamaIndex, or equivalents, especially for managing stateful, multi-turn AI agents.
• Experience with evaluation and monitoring tools including DeepEval, LangSmith, Langfuse, and Arize Phoenix for testing and observability.
• Strong MLOps skills encompassing Docker multi-stage builds and GitLab CI pipeline automation for testing, linting, and security scanning.
• Academic background holding a Bachelor's degree in STEM fields like Computer Science, Engineering, Mathematics, or Physics.
• Advanced Python skills including data validation with Pydantic and asynchronous REST API development using FastAPI.
• Hands-on knowledge of retrieval engineering with vector databases such as Vertex AI Vector Search, Weaviate, Milvus, or pgvector, and understanding of indexing and chunking strategies for hybrid dense/sparse search.
• Experience with security-related testing frameworks (e.g., Giskard, PyRIT) to detect LLM vulnerabilities and perform adversarial assessments.
• Preferred background in fintech sectors such as payments, banking, or financial services.
• Certifications like Google Professional Machine Learning Engineer or AWS Certified Machine Learning - Specialty are highly valued.

Eligibility

Open to candidates meeting the above qualifications and interested in advancing AI applications in fintech.