At Susquehanna International Group (SIG), data is the fundamental driver of every trading decision, quantitative model, and risk assessment. As a Data Engineer, you do not simply move data from one repository to another; you design, build, and optimize the high-performance pipelines that process massive volumes of market and proprietary data. Your work directly impacts the speed and accuracy with which SIG can identify market opportunities and execute trades in highly competitive global markets.

In this role, you will collaborate closely with quantitative researchers, traders, and software developers to translate complex financial data into clean, accessible, and highly optimized datasets. The systems you build must handle extreme scale and maintain near-zero latency, requiring a deep understanding of hardware limitations, memory management, and algorithmic efficiency. Whether you are optimizing data ingestion frameworks or refining historical databases for backtesting, your contributions are critical to maintaining SIG's competitive edge.

To succeed as a Data Engineer at SIG, you must possess a passion for solving intricate performance bottlenecks and a desire to understand the inner workings of your tools. This is an environment that rewards technical curiosity, analytical rigor, and a proactive approach to engineering challenges. If you thrive on building robust systems where every millisecond and byte of memory matters, this position offers an incredibly rewarding intellectual and professional playground.

To help you prepare effectively, we have compiled representative questions based on real candidate experiences at Susquehanna International Group. These questions highlight the core competencies evaluated during the interview process, specifically focusing on advanced Python programming, deep library knowledge, and data engineering fundamentals.

Python & Core Programming

This category tests your mastery of Python beyond basic syntax, focusing on language internals, memory management, and advanced programming paradigms.

• Explain how Python decorators work under the hood and write a custom decorator to log the execution time of a function.
• What is the difference between a generator and a list comprehension in terms of memory utilization and execution?

Preparing for an interview at Susquehanna International Group requires a shift in mindset from general software development to high-performance data manipulation. You should focus on demonstrating not just that you can write code that works, but that you understand exactly how that code executes at the system level.

Technical Rigor – You must demonstrate a granular understanding of your primary tools. Do not just memorize APIs; know how Python and Pandas manage memory, how data structures are laid out, and how to write highly optimized code.

Problem-Solving & Efficiency – Interviewers will evaluate how you approach complex, ambiguous data challenges. They want to see you break down a problem systematically, identify performance bottlenecks early, and propose highly efficient solutions.

Collaboration & Communication – Because Data Engineers work closely with trading desks and quantitative researchers, you must be able to explain complex technical trade-offs clearly. Show that you can translate business requirements into technical specifications seamlessly.

The interview process for a Data Engineer at Susquehanna International Group is structured to thoroughly evaluate both your technical depth and your alignment with the team's collaborative culture. The process is rigorous but highly structured, typically progressing from initial technical assessments to intensive, multi-round technical evaluations.

The journey begins with an initial phone screen or technical assessment focusing on core programming concepts, specifically in Python. If you pass this stage, you will move on to dedicated technical interviews that dive deep into data manipulation with Pandas and advanced Python paradigms. These rounds are designed to test your hands-on coding ability and your understanding of data structures.

The final stage is an intensive loop—often conducted on-site at offices like Dublin, Ireland or Bala Cynwyd, Pennsylvania. This final stage consists of multiple consecutive rounds with engineering team members, technical leaders, and sometimes security or infrastructure partners. These conversations focus on system design, pipeline architecture, and your ability to solve real-world data scaling challenges in real time.

The timeline shown above represents the typical progression for engineering candidates. You should use this visual structure to pace your preparation, ensuring you master core language fundamentals before moving on to complex system architecture and live coding scenarios.

To succeed at Susquehanna International Group, you must perform exceptionally well across several core technical domains. Below is a detailed breakdown of the primary areas where you will be evaluated.

Advanced Python Internals

SIG relies heavily on Python for data engineering and quantitative research pipelines. You must demonstrate an expert-level command of the language's internal mechanics and advanced features.

Be ready to go over:

• Decorators and Closures – Understanding how to implement, chain, and pass arguments to decorators, as well as how closures retain state.
• Generators and Iterators – Creating memory-efficient data streams using generators and understanding the iterator protocol.
• Memory Management – How reference counting, garbage collection, and object mutability affect application performance and memory footprints.
• Advanced concepts (less common) – Metaclasses, dunder methods, and utilizing the multiprocessing module to bypass the GIL.

Example scenarios:

• Implementing a custom decorator that caches function results (memoization) with an expiration timeout.
• Rewriting a memory-intensive file processing script to use generators instead of loading entire datasets into memory.

Pandas & Data Manipulation

Data manipulation is at the core of the Data Engineer role. You must show that you can clean, transform, and analyze large datasets rapidly and efficiently.

Be ready to go over:

• Under the Hood Mechanics – How Pandas utilizes NumPy arrays, memory alignment, and the difference between views and copies.
• Performance Optimization – Replacing slow iterative loops with vectorized operations, using .apply() efficiently, and leveraging categoricals.
• Data Alignment and Merging – Handling complex joins, multi-indexing, and aligning mismatched time-series datasets.
• Advanced concepts (less common) – Optimizing memory footprints using specific dtypes (e.g., float32 vs float64) and chunking massive CSV/parquet files.

Example scenarios:

• Optimizing a dataframe merge operation that is running out of memory on a standard developer workstation.
• Identifying and resolving a SettingWithCopyWarning in a complex data cleaning pipeline.