Senior AI Engineer - HCM - Full time

XNO is a quantitative AI technology company building institutional-grade financial infrastructure for professional investors across Vietnam and Southeast Asia. We operate at the intersection of systematic trading, AI, and financial data — serving fund managers, trading desks, and wealth professionals with tools they actually depend on for live decisions.

We’re looking for a Senior AI Engineer to own the architecture of the systems that advise real investment decisions on live financial data. This is not a role that takes a spec and implements it — you define how the agents think, set the technical bar for the team, and are accountable for correctness, reliability, and knowing when a model is wrong before it reaches a user.

What you’ll do

• Own the architecture of the multi-agent orchestration system end to end: intent classification, tool routing, sub-agent coordination, and response synthesis across complex financial queries
• Design and evolve the memory architecture: user profile memory, session context, and long-term behavioral memory across conversations
• Set the design standards for the tool library agents call — portfolio analyzer, market data fetcher, strategy backtester, scenario modeler, screener, report generator, compliance checker — and review other engineers’ additions to it
• Design and tune multi-tier LLM routing, optimizing quality, latency, and cost simultaneously across model tiers, with clear tradeoff criteria others on the team can follow
• Architect RAG pipelines for financial documents: embedding strategy, vector search, retrieval quality evaluation, and grounded, source-cited responses at production scale
• Integrate quantitative signals from the research team into agent reasoning — alpha factors, regime indicators, portfolio optimization outputs
• Build and own the LLM evaluation framework: golden-set regression tests, hallucination detection, factual accuracy scoring against verified financial data — and define what “good enough to ship” means for the team
• Build the observability stack for AI systems: token usage, latency per model tier, cost per conversation, and audit logs for compliance
• Set technical direction for the AI engineering function, review architecture decisions, and mentor other engineers on production LLM system design

Requirements

• 4+ years building production AI or LLM-powered systems — shipped to real users at scale, not research projects
• Track record owning the architecture of a non-trivial AI system from design through production, including the tradeoffs behind it
• Deep, hands-on experience with LLM APIs in production: Claude, OpenAI, or Gemini — streaming, tool use, structured outputs, context window management
• Agentic framework experience: LangChain, LangGraph, or equivalent — and the judgment to know when to build custom instead
• Proven RAG pipeline experience: embedding models, vector databases, retrieval quality evaluation at production scale
• Strong Python — async programming, clean architecture, testable code — and the ability to set code standards for others
• Production mindset: you instrument everything, handle failures gracefully, and catch model errors before users do
• Comfortable being the final technical decision-maker on AI architecture with minimal oversight

Bonus points

• Experience in fintech or financial services AI: compliance constraints, explainability requirements, data sensitivity
• Knowledge of Vietnamese financial products: mutual funds, VN30F derivatives, equities, bonds
• Multi-agent system design at scale: agent handoffs, tool orchestration, parallel sub-agent execution
• LLM evaluation and red-teaming for factual accuracy in high-stakes domains
• Familiarity with financial data sources: market data feeds, fund NAV series, macro indicators
• Prior experience with goal-based financial planning or robo-advisory systems
• Experience mentoring engineers or leading technical design reviews

Benefits

• Competitive salary + performance bonus
• Full ownership of AI architecture decisions — you define how the agents think
• Direct collaboration with PhD-level quant researchers and institutional finance professionals on live systems
• Work on AI systems where domain depth is the moat — not just prompt engineering
• Clear growth path to AI Architect or Head of AI Engineering as XNO scales across SEA