Research Institute for Federated Learning
Part-time Engineer (Temporary Appointment)
(Ref. 261009012)
[Appointment period: three to twelve months]
Duties
The appointee will be required to work for one of the constituent research units – Research Institute for Federated Learning (RIFL) under the PolyU Academy for Artificial Intelligence (PAAI). He/She will be required to:
(a) design and implement robust Large Language Model (LLM)-powered agent systems in specific domains, be capable of data collection and processing, model inference, model fine-tuning, self-evolution and long-term planning;
(b) integrate and optimise advanced autonomous coding and browser-control tools, such as Claude Code and OpenClaw, to automate internal developer workflows and external data-gathering tasks;
(c) fine-tune and prompt-engineer various LLMs, such as Claude 3.5, GPT-4 and Llama 3, to achieve specific agentic behaviors, with a focus on reducing latency and improving accuracy and efficiency;
(d) build the backend services, vector databases and state-management systems required to maintain agent "memory" across complex sessions;
(e) develop rigorous testing frameworks to measure agent reliability, safety and performance, moving beyond simple accuracy to evaluate "reasoning" paths;
(f) build a federated agentic system incorporating privacy-enhanced techniques to provide private and collaborative solutions to the agentic workflow; and
(g) perform any other duties as assigned by the Director of RIFL under PAAI or their delegates.
Qualifications
Applicant should:
(a) have a recognised degree in Computer Science, Artificial Intelligence, Data Science, Electronic Engineering or a related discipline;
(b) have at least five years of relevant post-qualification work experience;
(c) have in-depth knowledge in working with LLM APIs, such as Anthropic and OpenAI, and local deployments;
(d) have proven experience in agentic frameworks such as Claude Code, OpenClaw or Cursor;
(e) be proficient in coding and Machine Learning tools, such as Python and PyTorch;
(f) have experience in data and infrastructure with vector stores and containerization, such as Docker and Kubernetes, and GPU infrastructure;
(g) have a good command of both written and spoken English and Chinese; and
(h) have good communication skills.
Preference will be given to those with experience in federated learning and privacy computing.
Shortlisted candidates will be invited to sit for a written test.
Conditions of Service
The remuneration rate is HK$220 per hour.
Consideration of applications will commence on 16 October 2026 until the position is filled.
Posting date: 9 October 2026