PayPal logo

Machine Learning Engineer

PayPal

Bangalore
Freshers
Full-time
₹12 – ₹20 LPA
Posted 17 hrs ago
PayPal Banner

PayPal is hiring a Machine Learning Engineer for its Fraud team in Bangalore, covering the Crypto and Debit Card portfolios. You will work alongside senior ML engineers, data scientists and fraud strategy partners to build models that spot suspicious activity in real-time transactions. The day-to-day work covers data preparation, feature engineering, model training, evaluation and production monitoring. Every model you help improve has a direct effect on how safe it is for customers to buy, sell, transfer and spend. The role suits early-career engineers who want hands-on experience with production ML systems instead of purely academic projects.


This role puts you inside one of the most demanding applied machine learning environments in fintech. Fraud models at PayPal make decisions in real time on live money movement, so accuracy, speed and responsibility all matter from day one.


Why This Role Matters

Fraud detection is different from most ML work because the people on the other side keep changing their behaviour. A model that performs well this month may weaken next month as fraud patterns shift. In this team you will support models that protect crypto buy, sell and transfer activity as well as debit card transactions, two areas where speed and precision directly affect both customer trust and financial losses.

2
Fraud Portfolios: Crypto and Debit Card

You will not be building everything alone. The posting makes clear that you will work under the guidance of senior engineers, which makes this a good first step for someone who wants to learn how production fraud systems are designed and maintained.

Hybrid Full-Time Bangalore


A Typical Day In This Role

Expect your time to be split between data work, experimentation and collaboration. In the morning you might be cleaning a transactional dataset and checking it for gaps or inconsistencies. By afternoon you could be testing a new behavioural feature, such as how quickly a user makes repeated transfers, and comparing model results before and after adding it. You will also regularly talk with fraud operations partners to understand what new attack patterns they are seeing.

The team follows PayPal's hybrid model, which for most employees means three days in the office and two days at either the office or home.


Core Responsibilities Explained

The work breaks down into a few clear areas:

Model development. You will help build and improve models that detect fraud across debit card transactions and crypto activity.

Data preparation. You will preprocess, clean and analyze large transactional, behavioural and device datasets so the models learn from reliable inputs.

Feature building. Under senior guidance, you will create and test signals such as transaction velocity and behavioural patterns.

Evaluation. You will train and compare models using fraud-specific metrics, including precision, recall and false positive rates.

Deployment support. You will help with pipeline updates, testing and maintaining models in production.

Monitoring. You will watch deployed models for performance decline and drift, and flag problems to the team early.

Documentation. You will write clean, well-tested code and contribute to model documentation.


Skills Snapshot: Required vs Preferred

Area

Required

Preferred

Experience

1+ years, including internships, co-ops or academic projects

Fraud, risk, anomaly detection or payments exposure

Education

Bachelor's degree or equivalent combination of education and experience

Master's in CS, ML, Statistics or related field

ML frameworks

TensorFlow, PyTorch or scikit-learn

XGBoost, LightGBM, transformer architecture

Programming

Python and SQL

Git and software engineering fundamentals

Platforms

Not mandatory

AWS, Azure or GCP, plus Spark

Internships and academic projects count as experience


Technologies You Should Know

Python and SQL are the foundation of this role, so be comfortable writing queries on large tables and building data pipelines in Python. For modelling, you should be familiar with at least one major framework.

Python SQL scikit-learn TensorFlow PyTorch XGBoost LightGBM Spark Git

Gradient-boosted models are widely used in fraud detection because they handle tabular transaction data well, so spending time on XGBoost or LightGBM before you apply is a sensible investment. Cloud exposure on AWS, Azure or GCP is a plus rather than a requirement.


What Recruiters May Evaluate

Fraud is an imbalanced problem, since genuine transactions vastly outnumber fraudulent ones. Interviewers will likely check whether you understand why plain accuracy can mislead and how precision, recall and false positive rates tell a more honest story. Expect questions on:

Handling imbalanced data and choosing the right evaluation metric
Feature ideas for detecting unusual spending or transfer behaviour
Basic statistics, including distributions, sampling and hypothesis testing
SQL queries involving joins, aggregations and window functions
How you would notice and respond when a model's performance begins to drift

In fraud modelling, understanding the problem and the right metric often matters more than picking the fanciest algorithm.

How The Hiring Journey May Look

PayPal does not publish a fixed process for this role, so the flow below is a typical pattern for ML roles and may differ in practice.

flowchart LR
A[Application] --> B[Resume Screening]
B --> C[Technical Rounds]
C --> D[Hiring Manager Round]
D --> E[Offer]

Things Worth Learning Before Applying

Start with fraud-specific concepts: anomaly detection, class imbalance techniques such as resampling and cost-sensitive learning, and threshold selection. Then build a small end-to-end project, for example a credit card fraud detection model on a public dataset, with clear feature engineering and an honest evaluation. Also learn how to explain model drift, because monitoring deployed models is a stated part of this job. A basic understanding of how payments and crypto transactions flow will help you understand why certain signals matter.


What Can Strengthen Your Resume

Projects that show the full cycle carry more weight than notebooks that stop at training. Describe the data you worked with, the features you built, the metrics you chose and why, and what you improved. If you have internship or project experience in fraud, risk or payments, place it near the top. Keep your Git history clean and your code readable, since writing well-tested code is part of the job description.


Collaboration And Team Environment

You will work with data scientists, software engineers, senior ML engineers and fraud operations partners. Each group sees the problem differently: operations teams know the live attack patterns, engineers care about system reliability, and data scientists focus on model quality. Being able to explain your findings clearly to non-ML colleagues is a real advantage here.


Long-Term Career Direction

Starting in a fraud ML team builds a skill set that carries well into senior ML engineering, risk modelling and applied data science. As you take on more ownership of features, deployment and monitoring, you can move toward designing models independently. PayPal also lists senior and staff-level ML engineering openings for the same fraud domain, which suggests a visible growth path within the team.


Candidate Safety Note

PayPal states that it does not charge candidates any fees for courses, applications, resume reviews, interviews, background checks or onboarding. It also says it will never ask for passwords, one-time passcodes or verification codes, and that official communication comes only from PayPal email domains. Treat any request that breaks these points as a scam.


Keywords for Resume

Machine Learning • Fraud Detection • Python • SQL • scikit-learn • TensorFlow • PyTorch • XGBoost • LightGBM • Feature Engineering • Anomaly Detection • Model Evaluation • Precision and Recall • Data Preprocessing • Model Monitoring • Spark • AWS • Git • Statistics • Payments • Risk Analytics


Final Perspective

This role offers a rare chance to learn ML engineering on live, high-stakes systems with senior guidance close at hand. Candidates with solid Python and SQL skills, a good grip on model evaluation and a genuine interest in fraud and payments will find it a strong foundation for a long-term ML career.


The above article is written by me, a person interested in technology, automobiles, modern gadgets, movies, music, and clean aesthetics.

Disclaimer

This job listing is shared for informational purposes only. We are not affiliated with the hiring company. All applications must be submitted through the official company website.

Recent Postings

JJOBS