About Client
My client is a leading digital asset trading firm and liquidity provider that aims to deliver exceptional risk-adjusted returns. With the combination of comprehensive mathematical analysis, extensive financial market knowledge, and cutting-edge technology solutions, their trading model has stood the test of time. They are the avant-garde within the market space in terms of systematic decision making, algorithmic execution and active risk management.
Responsibilities
- Implement LLM agents in Python (tool use, function calling, orchestration, guardrails, evaluation loops).
- Build and maintain MCP servers/clients that securely connect StarRocks and MySQL, market data feeds, and back-testing/risk analytics engines to LLMs.
- Design and harden tool-invocation workflows so models can safely fetch data and compute results with auditing and least-privilege access.
- Wrap financial/market databases (StarRocks, MySQL) with read-only, aggregated APIs for consistent model consumption.
- Collaborate with quant researchers and engineers to integrate back-testing, risk, and analytics services into agent workflows.
- Contribute to secure, scalable AI architectures with logging, cost telemetry, and optimization in mind.
- Stay current with the LLM and MCP ecosystem evaluate new tools, frameworks, and models.
Requirements
- Currently pursuing or recently completed a degree in Computer Science, Data Science, Engineering, or a related field.
- Strong programming skills in Python (typing/pydantic, asyncio, packaging, testing).
- Very familiar with Linux (shell, processes, networking basics, permissions comfort with debugging tools and scripting).
- Familiarity with LLM APIs and frameworks (e.g., OpenAI, Anthropic, LangChain/Guidance/LlamaIndex or similar).
- Good skills using GPT, GPT Codex, Claude Code, or GitHub Copilot to speed up development while maintaining code quality.
- Basic understanding of the Model Context Protocol (MCP) or strong willingness to learn quickly.
- Working knowledge of StarRocks and MySQL data modeling and query performance ability to design safe read-only access patterns.
- Experience with backend development in Python (FastAPI/Flask/Django) designing clean REST/WebSocket endpoints.
- Understanding of containers (Docker) Kubernetes familiarity is a plus.
- Strong problem-solving skills, curiosity, and eagerness to learn.
- Solid communication skills comfortable collaborating with cross-functional teams.
Nice to Have
- Exposure to financial market data, backtesting pipelines, or risk/analytics systems.
- Experience with observability (structured logging, tracing, metrics) and cost controls for LLM workloads.
- Familiarity with authN/authZ, API keys, and secrets management.
.Kindly note that only shortlisted candidates will be notified. Thank you.
Dylan Choo
Rewarding Talents International
EA License: 22S1387
EA Personnel: R2094227
