AI Engineer

About Treasury Cube

Treasury Cube is a modern Treasury Management System (TMS) helping corporate treasury teams take control of cash, payments, bank connectivity, in-house banking, and FX exposure. We're building a platform that turns the daily grind of treasury operations, bank statement processing, cash positioning, intercompany settlements, payment workflows into something faster, smarter, and decision ready.

We're now embedding AI deeply into the platform to give treasurers the kind of insight, automation, and conversational access that legacy TMS vendors simply can't deliver. This is where you come in.

The Role

We're looking for an AI Engineer to lead the design and delivery of AI capabilities across Treasury Cube. You'll own the AI engineering stack end-to-end from model orchestration on Azure AI , to retrieval pipelines over treasury data, to the production systems that put insights and automation in front of CFOs and treasurers every day.

This is a hands-on, build-from-the-ground-up role. You'll shape architecture, set patterns, and ship features customers actually use within weeks, not quarters.

What You'll Build

• AI Insights Engine: Generate explanatory commentary and recommendations on cash positions, FX exposures, payment anomalies, forecast variances, and intercompany flows. Think: turning a 12-bank, 8-currency cash dashboard into a one-paragraph "here's what changed and why" briefing.
• AI Chatbox / Conversational Treasury Assistant: A natural-language interface where users can ask "what's my USD position next Friday?" or "show me payments held up in approval over $500K" and get accurate, grounded answers backed by the underlying treasury data.
• AI-Based Task & Notification Automation: Intelligent routing, prioritization, and summarization of alerts (covenant breaches, sweep failures, payment exceptions, hedge effectiveness drift). Auto-generated tasks with context, suggested actions, and assignment logic.
• Monitoring & Observability for AI Systems : Token usage, latency, hallucination detection, prompt regression, cost-per-customer tracking, and human-in-the-loop feedback pipelines. If we can't measure it, we won't trust it in front of treasurers.
  

What We're Looking For

Must-haves:

• 3+ years of software engineering experience, with at least 1–2 years building production LLM or AI applications
• Strong, hands-on experience with the Azure AI stack Azure OpenAI Service, Azure AI Foundry, Azure AI Search (vector + hybrid retrieval), Azure AI Document Intelligence, and Azure Content Safety
• Solid grasp of RAG architectures , prompt engineering, function/tool calling, structured outputs, and evaluation frameworks
• Experience designing agentic workflows and multi-step automation (Azure Functions, Logic Apps, Durable Functions, or equivalent)
• Strong backend engineering fundamentals Python and/or C#/.NET, REST APIs, async patterns, SQL
• Production mindset: you think about cost, latency, observability, prompt versioning, and graceful degradation from day one

Nice to have:

• Experience in FinTech, Treasury, ERP, or financial data domains
• Familiarity with ISO 20022 (CAMT, PAIN), SWIFT, or banking integrations
• Background in MLOps / LLMOps (Azure ML, Prompt Flow, model evaluation pipelines)
• Hands-on with Semantic Kernel, LangChain, or LlamaIndex
• Experience with multi-tenant SaaS architectures and data isolation patterns

Why Join Treasury Cube

• Real ownership. You'll be the AI lead, not the third engineer on a team of fifteen. Architecture decisions you make this quarter will shape the platform for years.
• A domain that rewards intelligence. Treasury data is structured, high-stakes, and underserved by AI. The problems are genuinely hard, and the wins are visible to CFOs the day you ship.
• Fast feedback loops. Small customer base today (early-stage), which means you'll talk to actual users, see your features in production within weeks, and iterate based on real treasurer feedback.
• Work that compounds. Every insight, automation, and chatbox capability you ship becomes part of a platform that scales across customers not a one-off engagement.

How to Apply

Send your CV, GitHub or portfolio (if relevant), and a short note on one production AI system you've built what it does, what you'd change, and why to HR@treasurycube.com .

Bonus points if you include thoughts on how you'd approach building a treasury AI assistant that earns the trust of a CFO.