【Company Profile】
HIGGS HK TECHNOLOGY LIMITED is a leading provider of artificial intelligence solutions in the financial trading industry. In the ever-changing capital markets, the greatest challenge lies in accurately forecasting market trends; traditional models often fail to capture market complexity, resulting in missed trading opportunities and increased risk. We are committed to overcoming these challenges by building high-performance, AI-driven infrastructure for China’s financial sector. Our AI solutions enable rigorous traders to effectively manage various models and datasets within a regulatory framework, allowing them to focus on deeper research and make more informed trading decisions, ultimately achieving outstanding returns. Our core values are: Uphold Integrity, Innovate with Insight, Maintain Humility, and Act Altruistically. We have branches in Kuala Lumpur and Ho Chi Minh City, forming an international development team.
【Position Overview】
We are seeking a CNN-LSTM Model Engineer with solid experience in deep learning and time series modeling. You will primarily be responsible for developing and optimizing deep learning models based on CNN-LSTM for complex tasks such as price forecasting and sentiment analysis. In addition to a strong background in machine learning and natural language processing, candidates should have an in-depth understanding of traditional time series models—particularly ARIMA, SARIMA, GARCH, and their limitations—so they can integrate cutting-edge deep learning techniques (e.g., FinBERT, TextBlob for text sentiment analysis) with traditional time series forecasting methods to provide data and model support for real trading strategies.
【Key Responsibilities】
Deep Learning Model Development & Optimization
Design, develop, and maintain CNN-LSTM and other hybrid deep learning architectures for predicting and analyzing financial markets (e.g., salmon spot prices).
Compare and contrast LSTM, GRU, CNN, MLP, etc., continuously optimizing model structures and hyperparameters (kernel size, LSTM units, learning rate, dropout, etc.).
Evaluate models using MSE, RMSE, MAE, MAPE, and employ Diebold-Mariano (DM) tests to verify the statistical significance of performance differences.
Time Series Analysis & Traditional Model Insights
Expertly handle ARIMA, SARIMA, GARCH, and other traditional time series models, understanding their assumptions, applicable scenarios, and limitations in nonlinear, volatile environments.
Benchmark and compare these traditional models against deep learning approaches to support final decisions and strategy deployment.
Demonstrate deep knowledge of decomposition, stationarity checks, differencing, and seasonal adjustments in time series.
Sentiment Analysis & Multivariate Modeling
Combine historical price data with news text to build multivariate time series models, capturing the potential impact of sentiment on price trends.
Use FinBERT, TextBlob, and other sentiment analysis tools to extract and quantify textual sentiment features; explore how integrating multi-source data improves forecasting performance.
Familiarity with data preprocessing, feature extraction, and normalization to ensure data quality and consistency.
Technical Innovation & System Integration
Keep abreast of cutting-edge techniques in deep learning and natural language processing, applying the latest research findings to trading systems and quantitative strategies.
Collaborate with data engineering and operations teams to design large-scale, distributed systems ensuring high concurrency and low latency in data flows.
Deploy and maintain deep learning models in production environments, monitoring model bias, quickly identifying and resolving anomalies.
Cross-Team Collaboration & Technical Support
Work closely with quantitative researchers and trading strategy teams to interpret model outputs and provide explainable insights.
Draft, maintain, and refine technical documentation on model design, data processes, and key algorithms; deliver training and support to internal tech teams and external partners.
【Qualifications】
Education & Background
Bachelor’s degree or higher in Computer Science, Statistics, Mathematics, Financial Engineering, or related fields.
Prior project experience in the financial or quantitative trading sectors is a plus.
Technical & Modeling Expertise
Proficient in deep learning frameworks such as TensorFlow or PyTorch, with hands-on experience in CNN and RNN (LSTM/GRU).
Familiar with CNN-LSTM hybrid models for time series, with a clear understanding of both their strengths and limitations.
Solid grasp of ARIMA, SARIMA, GARCH theory and applicability, and the ability to articulate why these traditional models are limited in highly volatile nonlinear markets.
Experienced in using FinBERT, TextBlob, or other NLP/sentiment analysis tools, especially in integrating textual sentiment into regression or forecasting models.
Programming & Tooling
Advanced proficiency in Python and data analytics libraries (NumPy, Pandas, Scikit-learn), with end-to-end data preprocessing and feature engineering capabilities.
Familiarity with Docker, Git, and other DevOps tools; knowledge of common databases (SQL/NoSQL) or message queues (Kafka) is preferred.
Ability to produce high-quality, reproducible code and maintain relevant documentation.
General Skills
Strong logical thinking and problem-solving abilities for identifying and resolving challenges in models or systems.
Excellent teamwork and communication skills for close collaboration with quantitative trading, data engineering, and operations teams.
A proactive learning mindset and innovative spirit, staying highly enthusiastic about emerging technologies.
Interest in or knowledge of financial markets (especially commodities and spot markets) is highly desirable.
【Compensation & Benefits】
Competitive base salary and performance bonuses.
Flat organizational structure with a positive, dynamic team culture.
Flexible work arrangements with up to 100% remote options, plus an annual overseas work allowance.
Multiple company-sponsored overseas trips each year, with various recreational activities (sports, board games, etc.).
Extensive opportunities for career development, continuous learning, and experimentation with cutting-edge technology.
【Work Locations】
Kuala Lumpur, Malaysia & Ho Chi Minh City, Vietnam