The Team 

Applied AI is the initiative to make Systematic, Man Group's quantitative investment business, an AI-native research firm: making general-purpose AI useful for quantitative research, and closing the gap between what frontier models can do and what the research process needs. We sit centrally within it as a combined research and technology team, and collaborate with Systematic's research teams across London, Boston, Sofia, Hong Kong and Shanghai to integrate AI into their research processes. 
 
Applied AI is a newly established team, and much of what we build is greenfield. Within the initiative, our technologists own the AI-native foundations: the knowledge layer over our code, research and data; the platform for long-horizon multi-agent research systems; and the evaluation harnesses and patterns that let researchers build reliable agent systems themselves. 

The team is part of Systematic’s broader technology group of over ninety technologists, with backgrounds spanning Computer Science, Mathematics, Physics and Engineering. What unifies us is a passion for technology and high-quality code, and a conviction that AI will change how systematic research is done. We are small and hands-on: everyone on the team writes code, works with researchers and owns what they ship. 

Role Responsibilities 

• Work collaboratively with researchers to build and own the AI-native research process, scoping problems and contributing ideas as well as code 

• Design, build and maintain agents and agentic workflows for research 

• Build knowledge bases and retrieval systems that give agents and Quant Research reliable access to the firm’s research, code and data 

• Build evaluations and benchmarks for AI tools and workflows, so that their value is measurable and changes to models, prompts or tooling can be compared 

• Follow developments in applied AI, prototype the ones that matter quickly, and harden what works into the platform 
  

Key Competencies 

Essential 

• Strong software engineering in Python; experience building production applications and services in a formal development environment 

• Proficient on Linux and with the Python numerical stack 

• Strong interpersonal skills; able to work closely with quantitative researchers, traders and senior business people alike 

• Experience building agents, agentic workflows, and harnesses 

• A solid grounding in machine learning and modern LLMs: their strengths, weaknesses, limitations and evolution 

• Good judgement about what to build; able to turn an ambiguous need into a useful tool 

• Proficient, ambitious use of agentic coding tools such as Claude Code or Codex 

• Demonstrable passion for applied AI and technology e.g. personal projects, open-source involvement, writing or talks 

 
Advantageous 

• Hands-on experience of quantitative finance or of working alongside quants 

• Experience building knowledge bases and retrieval systems 

• Hands-on experience of reinforcement learning or LLM post-training, e.g. fine-tuning, RLHF, reward modelling 

• Awareness of frontier AI research and products, in particular long-horizon agents and recursive self-improvement 

• Experience with distributed compute frameworks like Spark, Ray or Kubernetes 

• A research background, academic or industrial, in any field