AI Engineer

AXG (Solowin Holdings); Hong Kong Island; On-site; HKD 1,000,000 - 1,800,000; Full time

Job summary

We are seeking an expert in the engineering and practical deployment of large language models (LLMs) and generative AI for financial applications. The role focuses on transforming advanced AI capabilities into reliable financial services and risk control solutions.

Key areas include RAG-based Q&A, multi-step reasoning, and the development of financial decision-making agents. You will work on data privacy, compliance, and scalable model deployment for intelligent risk control and automated

Qualifications

• Bachelor’s degree or above with relevant NLP/AI experience.
• Hands-on experience with large models and practical implementation.
• Proven ability to deploy LLMs in real-world financial scenarios.
• Strong engineering skills in deep learning frameworks (e.g., PyTorch).
• Experience with RAG, SFT, and model optimization techniques.

Responsibilities

• Engineer the integration of LLMs for core financial use cases.
• Develop generative content workflows for investment research and trading.
• Design and implement Agentic Finance and multi-agent systems.
• Lead domain-specific fine-tuning and data privacy/compliance efforts.
• Optimize model inference latency and cost (distillation/quantization).

Skills

NLP

Multimodal data processing

Large model deployment

Algorithm design

Education

Bachelor’s degree in a relevant field

Tools

PyTorch

vLLM

sglang

Qwen series

Fine-tuning systems

Job description

We are looking for an expert in the engineering and practical implementation of large language models (LLMs) and generative AI for financial application scenarios. The role focuses on transforming cutting-edge AI capabilities into efficient, reliable financial services and risk control solutions. You will be responsible for the deep integration and deployment of LLMs and generative AI in financial business scenarios (such as intelligent risk control, quantitative investment, automated trading, and customer service), driving the digital and intelligent upgrade of traditional financial businesses. Key areas include: multi-step reasoning based on large language models, RAG-based financial Q&A and compliance review systems, design and implementation of financial decision-making agents (Agentic Finance), understanding of multimodal financial data (such as financial reports, market data, and public sentiment), as well as incremental pre-training, SFT, and interpretability research of large models tailored for financial scenarios.

KeyResponsibilities

1.Responsible for the engineering implementation of large language models in core financial scenarios, including integration of intelligent risk control models, construction of RAG systems for complex financial data, and multi-step reasoning enhancement.

2.Responsible for the research and development of generative financial content generation based on LLMs (such as investment research reports and code generation) and auxiliary models for automated trading strategies.

3.Participate in the core algorithm design and system implementation of Agentic Finance (financial decision-making agents) and multi-agent collaborative systems.

4.Responsible for SFT of domain-specific large models in finance, data construction (with a focus on data privacy and compliance), and the establishment of multi-dimensional evaluation systems (such as accuracy, stability, and interpretability).

5.Responsible for model engineering optimization to control inference latency and cost while ensuring performance (such as model distillation, quantization, and inference acceleration).

KeyRequirements

1.Bachelor’s degree or above, with more than 3 years of experience in NLP, multimodal, or recommendation/decision algorithms, including at least 1 year of hands-on experience and engineering practice with large models.

2.Proficient in the training and fine-tuning systems of mainstream large models such as the Qwen series, expert in methods like SFT, and able to adapt them to financial scenarios.

3.Experience in successfully deploying large language models in real-world complex business scenarios with clear business benefits.

4.Solid engineering capabilities, proficient in deep learning frameworks (such as PyTorch), skilled in efficient inference frameworks such as vLLM and sglang, and able to independently address high-performance requirements in model deployment.

5.Preferred qualifications: Contributions as a contributor to well-known open-source projects or frameworks; practical AI R&D or deployment experience at well-known companies.