Descrição da vaga
<div><strong>About Thunes</strong></div>
<div><strong><br></strong>Thunes is the Smart Superhighway to move money around the world. Thunes’ proprietary Direct Global Network allows Members to make payments in real-time in over 140 countries and more than 90 currencies. Thunes’ Network connects directly to over 12 billion mobile wallets, stablecoin wallets and bank accounts worldwide, as well as 15 billion cards via more than 220 different payment methods, such as GCash, M-Pesa, Airtel, MTN, Orange, JazzCash, Easypaisa, AliPay, WeChat Pay HK and many more.</div>
<div> </div>
<div>Thunes’ Direct Global Network differentiates itself through its worldwide reach, in-house SmartX Treasury System and Fortress Compliance Platform, ensuring Members of the Network receive unrivaled speed, control, visibility, protection, and cost efficiencies when making real-time payments, globally. Members of Thunes’ Direct Global Network include gig economy giants like Uber and Deliveroo, super-apps like Grab and WeChat, MTOs, fintechs, PSPs and banks.</div>
<div> </div>
<div>Headquartered in Singapore, Thunes has offices in 14 locations, including Barcelona, Beijing, Dubai, Hong Kong, Johannesburg, London, Manila, Nairobi, New York, Paris, Riyadh, San Francisco and Shanghai. For more information, visit: <a href="https://www.thunes.com/">https://www.thunes.com/</a></div>
<p><strong>Context of the role</strong></p>
<p>We are looking for a highly driven, self-motivated, and technically excellent engineer who is truly excited about building the "AI Operating System" of a high-growth Fintech.</p>
<p>You will need to combine a startup mindset with the rigor of a regulated financial institution, moving beyond simple chatbots to build "Stateful" Agentic Systems and Multimodal RAG pipelines that can reason across text, charts, and financial tables. This role is responsible for developing our core generative capabilities and getting them to production in the most efficient, secure, and observable way possible. We architect solutions that bridge our AI tech stack with our enterprise data lake on AWS. We rely on our AI Engineers to possess "Deployment Autonomy", meaning you architect, implement, test, and deploy your own solutions.</p>
<p><strong>Key Responsibilities</strong></p>
<ul>
<li>Architect production-grade RAG workflows on cloud resources (e.g. AWS, GCP) using reliable tools (e.g. LangGraph / LlamaIndex). Build agents that can plan multi-step financial tasks</li>
<li>Implement multimodal reasoning pipelines that ingest and reason over mixed-media financial documents (e.g. Charts, Graphs, Tables)</li>
<li>Act as the internal adversary (Red Teaming). Design cases to stress-test models against Prompt Injection, Jailbreaking and PII Leakage before they touch sensitive data</li>
<li>Define and implement the "Definition of Done" for AI using quantitative evaluation. Use evaluation tools (e.g. DeepEval) to score models on Faithfulness and Relevancy. Use these metrics to validate model upgrades and cost-optimisation experiments</li>
<li>Champion Deployment Autonomy: Containerise your solutions (e.g. Docker), define the infrastructure and deploy to our cloud resources via GitLab CI</li>
<li>Ensure governance by implementing deterministic guardrails (e.g. NeMo Guardrails, Guardrails AI) to enforce JSON output schemas and block non-compliant financial advice</li>
</ul>
<p><strong>Professional Experience and Qualifications</strong></p>
<ul>
<li>5+ years of total AI engineering experience, with 2+ years of hands-on experience building and deploying NLP or GenAI solutions in production</li>
<li>Multi-Cloud Fluency: Expertise in architecting solutions on major cloud platforms (e.g. AWS, GCP). Mastery of cloud native LLM ecosystems (e.g. Bedrock, Vertex AI). Strong operational grasp of cloud services (e.g. IAM, Storage)</li>
<li>Mastery of GenAI Orchestration: Expert level skills in LangChain or LangGraph (specifically for stateful, multi-turn agents) or LlamaIndex and other equivalent tools/libraries</li>
<li>Experience in Evaluation & Monitoring: Proficiency with open source tools for testing (e.g. DeepEval). Experience implementing full tracing and observability tools (e.g. LangSmith, Langfuse, Arize Phoenix)</li>
<li>Passionate about MLOps: Ability to write GitLab CI pipelines (automated testing, linting, scanning) and strong Docker skills (multi-stage builds)</li>
<li>Holding a Bachelor’s degree in a STEM field (Computer Science, Engineering, Mathematics, Physics)</li>
<li>Python Expertise: Proficiency with data validation (e.g. Pydantic) and asynchronous REST APIs (e.g. FastAPI)</li>
<li>Retrieval Engineering: Hands-on experience managing vector databases (e.g. Vertex AI Vector Search, Weaviate, Milvus or pgvector). Understanding of indexing and chunking strategies for hybrid search (Dense + Sparse)</li>
<li>Experience in Security & Red Teaming: Knowledge of adversarial testing frameworks (e.g. Giskard, PyRIT) to scan for LLM vulnerabilities</li>
<li>Interest in the Fintech Industry: Prior experience in Payments, Banking or Finance Services is strongly preferred</li>
<li>Certifications: Google Professional Machine Learning Engineer / AWS Certified Machine Learning - Specialty is highly preferred</li>
</ul>
<p><em>Sound like you? Apply now!</em></p>
<p> </p>