Data Scientist

Paragon Credit Alliance 604 Corporation; Makati; In loco; PHP 650.000 - 900.000; Tempo pieno

Descrizione del lavoro

Paragon Credit Alliance 604 Corporation is seeking a Data Scientist to help develop credit risk models, scorecards, and portfolio analytics that drive lending and portfolio decisions. The role targets candidates with 2–3 years of experience who are proficient in Python and SQL, and have a solid grounding in statistics and machine learning.

You will work with customer, financial, and lending data, build and validate models, and support monitoring and ongoing improvements to risk metrics.

Competenze

• Around 2–3 years of experience in data analytics, data science, risk analytics, or a similar role.
• Good working knowledge of Python, particularly pandas, NumPy, and scikit-learn.
• Comfortable working with SQL.
• Strong foundation in statistics and machine learning.
• Experience actually building and evaluating models.
• Able to work independently and explain technical concepts clearly.

Mansioni

• Work with customer, financial, transaction, and lending data to understand borrower behavior.
• Clean, prepare, and analyze data using Python and SQL.
• Explore datasets and identify patterns for predicting credit risk.
• Help build and test statistical and machine learning models.
• Support the development of credit scores, risk segments, and scorecards.
• Test different modeling approaches and evaluate performance.
• Assist with feature engineering, model validation, and monitoring.
• Investigate data quality issues, leakage, bias, and shifts in performance over time.
• Analyze lending portfolio across segments, industries, products, and risk levels.
• Monitor portfolio performance including delinquencies, defaults, exposures, and losses.
• Identify trends and impact of changes in underwriting on portfolio risk.
• Help translate findings into practical recommendations for growth and risk management.
• Present findings clearly to technical and non-technical stakeholders.
• Connect borrower-level analyses to portfolio performance and business strategy.

Conoscenze

Data analytics

Python

SQL

Statistics

Machine learning

Problem solving

Independent work

Communication

Formazione

Bachelor's degree in Statistics, Mathematics, Economics, Computer Science, Data Science, Engineering, Physics, or another quantitative field

Strumenti

Python

SQL

Pandas

NumPy

scikit-learn

Descrizione del lavoro

Data Scientist

The role will focus on helping develop credit risk models, scorecards, and portfolio analytics that can support better lending and portfolio management decisions.

We're looking for someone with around 2–3 years of experience who is comfortable working with data and Python, has a good foundation in statistics and machine learning, and is interested in applying these skills to real business problems.

What you'll be doing

1. Data & Modeling

• Work with customer, financial, transaction, and lending data to better understand borrower behavior.

• Clean, prepare, and analyze data using Python and SQL.

• Explore datasets and identify patterns, trends, and variables that could be useful for predicting credit risk.

• Help build and test statistical and machine learning models.

• Support the development of credit scores, risk segments, and scorecards.

• Test different modeling approaches and evaluate how well they perform.

• Help with feature engineering, model validation, and model monitoring.

• Investigate issues such as missing data, data leakage, bias, overfitting, and changes in model performance over time.

2. Portfolio Analytics

• Analyze the lending portfolio across different customer segments, industries, products, and risk levels.

• Monitor portfolio performance, including delinquencies, defaults, exposures, and losses.

• Identify trends or segments where credit risk may be increasing.

• Look at how changes in customer mix, underwriting policies, or approval criteria could affect overall portfolio risk.

• Help develop metrics and analyses for tracking portfolio health.

• Help answer questions like:

  • Which types of borrowers are performing better or worse than expected?

  • Where are we seeing higher concentrations of risk?

  • How is the portfolio changing as we grow?

  • What happens to portfolio risk if we change our approval criteria?

  • Are our risk models still performing well as the portfolio changes?
    

3. Business & Decision Support

• Work with business and risk teams to understand the lending process and how analytical models will actually be used.

• Help translate analytical findings into practical recommendations.

• Think about the trade-offs between growth, credit quality, and risk.

• Present findings clearly to both technical and non-technical stakeholders.

• Help connect individual borrower-level analysis to the bigger picture of portfolio performance and business strategy.

What we're looking for

Must Have

• Around 2–3 years of experience in data analytics, data science, risk analytics, or a similar role.

• Good working knowledge of Python, particularly pandas, NumPy, and scikit-learn.

• Comfortable working with SQL.

• Good foundation in statistics and machine learning.

• Experience actually building and evaluating models.

• Strong analytical and problem-solving skills.

• Comfortable working with messy or incomplete data.

• Able to work independently and figure things out when the problem isn't clearly defined.

• Good communication skills and the ability to explain technical concepts in a simple way.

• Bachelor's degree in Statistics, Mathematics, Economics, Computer Science, Data Science, Engineering, Physics, or another quantitative field.

Nice to have

• Experience in lending, credit risk, banking, fintech, insurance, or financial services.

• Experience working with credit or risk models.

• Familiarity with concepts such as probability of default, default rates, risk segmentation, scorecards, WoE/IV, and model monitoring.

• Experience with portfolio monitoring or portfolio analytics.

• Experience working with financial or transactional data.

• Experience working with imbalanced classification problems or time-based validation.