AI Builder

JOB DESCRIPTION

AI Builder

Exchange Platform | Full-Time | Hong Kong / Remote

ABOUT THE ROLE

We are looking for a hands-on AI Builder obsessed with automation to design, develop, and deploy intelligent systems that power our exchange platform. This is a builder role with a heavy automation mandate — you will architect and ship AI systems that eliminate manual processes, accelerate decision-making, and create self-running workflows across trading operations, compliance, customer experience, and internal tooling.

You will work at the intersection of agentic AI, large language models (LLMs), real-time data pipelines, and high-performance financial infrastructure. The ideal candidate doesn't just build AI features — they systematically identify every manual, repetitive, or slow process and replace it with something that runs itself.

KEY RESPONSIBILITIES

AI Automation Systems

• Identify and automate high-friction manual workflows across the exchange — from trade lifecycle operations and settlement reconciliation to customer onboarding, KYC/AML checks, and support ticket routing.
• Build multi-step AI Agent pipelines (using frameworks such as LangChain, CrewAI, AutoGen, or LangGraph) that autonomously execute complex, multi-decision workflows with minimal human intervention.
• Design event-driven automation triggers that respond to market conditions, user actions, or system alerts in real time — automatically initiating reports, escalations, or corrective actions.
• Automate compliance monitoring: build AI systems that continuously scan transactions, flag suspicious activity, generate regulatory reports, and surface anomalies for review — reducing manual compliance overhead by 80%+.
• Deploy automated market intelligence pipelines that ingest, parse, and summarise exchange data, order flow signals, and external market news without human curation.

LLM & Intelligent Interfaces

• Develop and maintain LLM-based pipelines for natural language interfaces, AI-powered user support bots, trading insights summarisation, and automated document generation (trade confirmations, reports, notices).
• Build RAG (Retrieval-Augmented Generation) systems grounded in exchange data, regulatory documents, and internal knowledge bases — enabling staff and users to query complex information instantly.
• Create intelligent notification and alerting systems that auto-generate contextualised, human-readable summaries from raw data streams.

Risk, Fraud & Operations Automation

• Build real-time anomaly detection and fraud flagging systems that automatically escalate or act on suspicious trading patterns, account behaviour, or data integrity issues.
• Automate risk reporting workflows: ingest position data, compute exposures, and generate structured risk summaries on a scheduled or trigger-based basis.
• Develop automated reconciliation agents that cross-check internal ledgers, exchange records, and external data sources — surfacing discrepancies without manual intervention.

Platform & Infrastructure

• Collaborate with product, engineering, and compliance teams to identify automation opportunities and translate them into scalable AI solutions.
• Monitor and evaluate automation performance in production; build feedback loops that continuously improve accuracy and reduce false positives.
• Ensure all automated AI outputs are auditable, explainable, and compliant with applicable regulatory standards.
• Stay current with developments in agentic AI, workflow automation, and financial AI; prototype and ship new automation approaches quickly.

REQUIREMENTS

Must-Have

• 3+ years of experience building and shipping production AI, ML, or automation systems.
• Strong proficiency in Python; experience with LLM and agent orchestration frameworks (e.g. LangChain, LangGraph, CrewAI, AutoGen, n8n, or similar).
• Hands-on experience with LLM APIs (OpenAI, Anthropic, Gemini, or open-source models via HuggingFace / Ollama).
• Solid understanding of RAG architectures, vector databases (e.g. Pinecone, Weaviate, pgvector), and embedding models.
• Experience building event-driven and workflow automation systems; familiarity with tools such as Zapier, Make, Prefect, Airflow, or custom agent pipelines.
• Experience deploying AI services in cloud environments (AWS, GCP, or Azure); comfortable with containerisation (Docker, Kubernetes).
• Strong systems-thinking mindset — able to map a business process end-to-end and identify every automation opportunity within it.
• Ability to work independently with high ownership — from ideation to production.

Nice-to-Have

• Background in fintech, crypto, or financial exchange platforms.
• Experience with real-time data pipelines (Kafka, Flink, or similar) and time-series databases.
• Familiarity with DeFi protocols, order book mechanics, or algorithmic trading systems.
• Experience with RPA (Robotic Process Automation) tools or low-code automation platforms.
• Experience fine-tuning or evaluating open-source LLMs.
• Knowledge of AI safety, output evaluation frameworks (e.g. RAGAS, DeepEval), and responsible AI practices.