Microsoft vs Goldman Sachs: Interview Question Comparison

Preparing for technical interviews requires understanding each company's specific focus. Microsoft and Goldman Sachs both test core algorithmic skills, but their approach, volume, and emphasis differ significantly. Microsoft's process is a marathon of classic coding problems, while Goldman Sachs focuses on a tighter set of questions with a stronger tilt toward financial and data processing logic. Your preparation strategy should reflect these differences.

Question Volume and Difficulty

The sheer volume of known questions is the most striking difference. Microsoft has approximately 1,352 tagged questions on CodeJeet, dwarfing Goldman Sachs' 270. This doesn't mean Microsoft asks more questions per interview, but it indicates a broader, more established public question bank from decades of engineering hiring.

The difficulty distribution also reveals distinct philosophies:

• Microsoft (E379/M762/H211): The curve peaks at Medium difficulty (56% of questions), with a substantial number of Easy (28%) and Hard (16%) problems. This reflects a balanced interview designed to assess strong fundamentals (Easy/Medium) with room to differentiate top candidates (Hard).
• Goldman Sachs (E51/M171/H48): The distribution is similar in shape but steeper, with Medium problems making up an even larger portion (63%). The Hard problem count is proportionally smaller. This suggests their technical screen is highly focused on reliably assessing competent, clean coding under typical constraints, with fewer "trick" questions.

In practice, a Microsoft interview is more likely to include a known, classic Hard problem (e.g., a complex DP or graph traversal), while a Goldman Sachs interview often stays within Medium territory but may require more careful handling of edge cases and data integrity.

Topic Overlap

Both companies heavily test the four foundational topics: Array, String, Hash Table, and Dynamic Programming. This is the core overlap you must master for either.

However, the context and application diverge:

• Microsoft questions often relate to system design, software architecture, and algorithmic optimization. Problems may involve designing data structures, manipulating text (simulating editors or compilers), or optimizing resource usage. The focus is on scalable, efficient computer science.
• Goldman Sachs questions frequently incorporate financial or quantitative logic. You might process time-series data (Arrays), clean and validate financial transaction strings, use Hash Tables to aggregate trade data, or apply DP to optimization problems like maximizing profit under constraints. The focus is on accuracy, data handling, and business logic.

# Example: A "profit" problem might appear at both.
# Microsoft: Framed as general array manipulation.
def max_profit(prices):
    min_price, max_profit = float('inf'), 0
    for price in prices:
        min_price = min(min_price, price)
        max_profit = max(max_profit, price - min_price)
    return max_profit

# Goldman Sachs: Might include transaction fees or multiple assets.
def max_profit_with_fee(prices, fee):
    cash, hold = 0, -prices[0]  # cash after sell, cash after buy
    for i in range(1, len(prices)):
        cash = max(cash, hold + prices[i] - fee)
        hold = max(hold, cash - prices[i])
    return cash

Which to Prepare for First

Prepare for Microsoft first. Its question bank is vast and covers the algorithmic fundamentals in greater breadth and depth. Mastering Microsoft's list, particularly the Medium and common Hard problems, will build a robust foundation that makes Goldman Sachs' focused list feel like a subset. The core topics are identical; the difference is context.

Your study path:

1. Grind the Core Topics: Achieve fluency in Array, String, Hash Table, and Dynamic Programming using Microsoft's problems. This builds muscle memory.
2. Practice Problem Translation: Learn to identify the underlying algorithmic pattern in a Goldman Sachs problem (e.g., realizing a "trade matching" problem is a variation of Two Sum using a Hash Table).
3. Add Financial Context: Once fundamentals are solid, practice applying them to quantitative scenarios—calculating returns, aggregating time-series data, handling bounds and validation.

Starting with Microsoft's broader set forces deeper understanding. Starting with Goldman Sachs' narrower set risks being underprepared for the range of problems Microsoft can ask. Build the general engine first, then adapt it to the specific domain.

For targeted practice, visit the company pages: Microsoft and Goldman Sachs.