About the Role
We are seeking top-tier fresh graduates / junior AI Scientists to build intelligent fraud detection and abuse prevention systems that secure our users and trading ecosystem.
This role targets high-caliber candidates with strong academic performance and proven Web3 internship experience, operating at the intersection of:
- Fraud & trading risk modeling
- AI Agent systems (LLM reasoning, tool use, multi-agent workflows)
- On-chain / trading data intelligence
You will help build production-grade AI and agent-driven systems to detect, explain, and prevent fraud, abuse, and manipulative trading behaviors in real time.
Key Risk Use Cases You'll Tackle
- Identity Risk: Fake KYC profiles, identity farming, synthetic user creation
- Account Takeover (ATO): Compromised credentials, anomalous login behavior, device takeover
- Toll Fraud & Drainers: Malicious contracts, phishing-induced approval scams
- Money Laundering: Cross-chain obfuscation, mixer usage, layering strategies
- Campaign Reward Abuse: Multi-account farming, airdrop exploitation, referral fraud
Trading-Related Abuse & Market Manipulation
- Wash trading / self-trading to inflate volume
- Spoofing / layering (fake order book depth)
- Abnormal high-frequency or latency arbitrage behaviors
- Coordinated trading rings / signal groups
- Bonus abuse via hedging / risk-free trading strategies
- Market manipulation patterns across accounts or venues
Responsibilities
AI / Machine Learning
- Build models for: anomaly detection (user & trading behavior) behavioral scoring (traders, accounts, entities) fraud and manipulation pattern discovery
- Engineer features from: on-chain transactions order book, trades, positions, PnL user activity (login, device, referral signals)
AI Agent Systems (Core Focus)
- Design LLM-powered AI Agents for: fraud investigation trading behavior analysis
- Implement: multi-step reasoning over user + trading data tool calling (risk systems, trading DB, on-chain APIs)
- Build multi-agent workflows (planner / analyzer / scorer) using: LangChain / LangGraph
- Develop RAG systems for: case analysis and investigation workflows
- Improve agent reliability: prompt design, evaluation, failure handling
Data & Engineering
- Build pipelines using: Python + PyTorchLLM stack (OpenAI / vLLM / LangChain / LangGraph)
- Develop systems for: real-time risk scoring monitoring and alerting
- Work cross-functionally to deploy production solutions
Collaboration & Learning
- Partner with trading, risk, and compliance teams
- Understand evolving fraud and market manipulation patterns
- Continuously improve models and agent systems
Qualifications
Must-Have (Strict Requirements)
- Master's in Computer Science, Data Science, Quantitative Finance, or related field
- GPA ≥ 4.0 / 5.0
- MANDATORY: Internship/working experience in crypto / Web3 industry (exchange, DeFi, on-chain analytics, trading platform, etc.)
- Strong foundation in: machine learning / statistics data analysis and feature engineering
- Hands-on experience with: Python + PyTorch real datasets (trading / user / transaction data preferred)
- Practical experience in at least ONE: fraud / risk / trading behavior analytics LLM / AI Agent systems (RAG, tool calling, workflows)
- Understanding of: trading data (orders, trades, positions) or blockchain data (transactions, wallets, contracts)
AI Agent Requirement (Key Differentiator)
- Hands-on experience building: LLM systems with RAG, tool use, multi-step reasoning
- Familiar with: LangChain / LangGraph or equivalent
- Able to explain: how agents plan, call tools, and handle complex workflows
Preferred
- Exposure to: trading systems, order book mechanics, or market microstructure
- Experience with: fraud detection / AML / abuse detection systems
- Knowledge of: clustering, graph analysis, or behavioral segmentation
- Built: dashboards, alerting systems, or internal risk tools
What We Offer
- Competitive compensation for top-tier junior talent
- Work on real trading + fraud risk systems at scale
- Direct exposure to AI Agents + Web3 + trading infrastructure
- Fast growth with high ownership
- Global team across AI, trading, fraud, and blockchain
