Job Description: Data Scientist

Position Purpose: We are seeking an experienced Machine Learning Data Scientist to develop and implement ML-based solutions for our Global Market activities. This role involves collaborating closely with trading, sales, structuring, and strategy teams to optimize decision-making, streamline processes, and anticipate trends using cutting-edge machine learning techniques.

Our applications of machine learning focus on:

- Automation: Streamline repetitive tasks, freeing team members to address more complex challenges.

- Process Optimization: Replace slow or complex automated processes with more efficient ML-driven solutions.

- Scalability: Utilize ML to assist in processing large volumes of information, enabling systematic, timely decisions.

- Prediction: Apply ML models to analyze large datasets for future trend forecasting, particularly in time series analysis.

Responsibilities:

- Investigate and analyze data from multiple sources to identify actionable insights.

- Conduct conceptual modeling, statistical analysis, predictive modeling, and optimization.

- Identify and address limitations in analytic models.

- Cleanse, normalize, and transform data for effective analysis.

- Develop hypotheses and validate them through rigorous experimentation.

- Extract embedded patterns and insights to guide informed business decisions.

- Design workflows for data extraction, transformation, and integration with existing systems.

- Ensure data integrity and uphold security standards.

- Maintain collaboration with global team members and provide support across time zones as needed.

Technical Skills to Evaluate:

- Machine Learning Techniques

- Gradient Descent / Gradient: Core optimization technique to minimize loss in training.

- Dimensionality Reduction: Techniques like PCA to simplify data and enhance model performance.

- Loss Function: Key metric for evaluating model predictions.

- Activation Function: Functions like ReLU, Sigmoid that introduce non-linearity into neural networks.

- Natural Language Processing (NLP)

- NLP Techniques: Proficiency in processing text data, including tokenization and syntactic parsing.

- NER (Named Entity Recognition) / Entity Extraction: Identifying and extracting entities from text.

- Few-Shot & Zero-Shot Learning: Techniques to perform tasks with minimal or no task-specific data.

- Transfer Learning: Applying knowledge from pre-trained models to new tasks.

Deep Learning Architectures:

- Transformers: Expertise in models like BERT, GPT for handling sequential data.

- Encoder & Decoder Models: Foundational elements of sequence-to-sequence tasks.

- Autoencoder: Model used for unsupervised learning, dimensionality reduction, and feature extraction.

- Attention Mechanism: Core concept in modern NLP, focusing on relevant parts of the input data.

- Large Language Models (LLM): Familiarity with models like ChatGPT, Mistral for advanced NLP tasks.

Technical Tools:

- Pytorch: Advanced knowledge in this deep learning framework.

- Optimization Techniques: Regularization, fine-tuning for model improvement.

Technical & Behavioral Competencies:

Qualifications: Bachelor's, Master's, or PhD in Computer Science, Data Science, or a related field.

Statistical Knowledge: Solid grounding in Probability Theory, Inference, and Linear Algebra.

Programming Skills: Proficiency in Python, NumPy, scikit-learn, pandas, TensorFlow, PyTorch, langchain.

IT Knowledge: Familiarity with operating systems, parallel processing, networking, and software engineering.

This role is ideal for professionals eager to deepen their impact in a global, dynamic market environment through the latest advancements in machine learning and AI.