← Back to Directory

Point72

Large Enterprises

Screna has 17 Point72 interview write-ups. The most reported round is the technical phone screen (9).

17 interview notes · updated Jun 2026

Point72·Software Engineer·Online Assessment (OA)

Jun 2026
SQL-heavy online assessment for a Data Engineer role at Point72, one problem but it was a beast. The question looked like a straightforward aggregation task until you got into the formatting requirements and started hitting weird PostgreSQL edge cases.
  • Given a cryptocurrency exchange transaction table, write a PostgreSQL query to detect potential fraud by finding sender-recipient pairs with 2 or more transactions, then aggregate results per sender showing: the time period of activity, and a formatted string of transaction amounts sorted descending (top 3 shown explicitly, the rest collapsed with a count suffix). Order final output by total amount descending.

“This one ate up way more time than it should have.”

View Post

Point72·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Interviewed for an ML Engineer role at Point72. One technical question about decoder model behavior, touched on sampling and temperature, interviewer seemed happy with the answer.
  • Given identical input, why would a decoder-based language model produce a different output on each run?

“Went straight to sampling and temperature, which is the obvious answer, but I made sure to actually explain the mechanism rather than just name-dropping the terms.”

View Post

Point72·Machine Learning Engineer·Technical Phone Screen

Jun 2026
Interviewed for an ML Engineer role at Point72. One technical question about transformer architecture, pretty short exchange from what I can tell.
  • What is the key difference between an encoder and a decoder in a transformer architecture?

“Mentioned the causal mask and that seemed to be the answer they were looking for.”

View Post

Point72·Software Engineer·Take-home Assignment

May 2026
Point72 Data Engineer take-home involving two PySpark problems around patient eligibility and medical billing data. Five functions to implement, each with its own gotcha. Pretty solid signal of what they actually care about on the job.
  • Set up a Spark session with the correct master configuration and application name.
  • Filter a medical bills table to only include records for patients who appear in the eligibility table.
  • Generate a full name column by combining first and last name fields from the eligibility table, and return the result in the required schema.
  • Find the member who paid the highest amount, given that the data was loaded from a CSV file.
  • Calculate the total amount paid across all records and return it as an integer scalar value.

“Straightforward but you'd be surprised how easy it is to fat-finger the master string.”

View Post

Point72·AI Engineer·Online Assessment (OA)

May 2026
Took an online assessment for an AI Engineer role at Point72. One section was a prompt engineering scenario with multiple sub-questions and a weird copy-paste restriction that I completely underestimated time-wise.
  • Design a prompt to filter news articles for an Olympic champion athlete, covering multiple specific sub-scenarios.

“Five sub-questions under one prompt engineering scenario.”

View Post

Point72·Software Engineer·Technical Phone Screen

May 2026
Interviewed for a quant engineer role at Point72. The technical questions were pretty focused on statistical fundamentals and how they scale, not the flashiest interview but it made me think harder than expected.
  • How do you estimate the coefficients in a linear regression model when you have far more observations than features?

“Said gradient descent first, which was fine, but then they asked if there's a closed-form solution and I kind of froze.”

View Post

Point72·Software Engineer·Technical Phone Screen

May 2026
Got a SQL problem for a Data Engineer role at Point72. One question, but it had enough layers to keep me busy for a while.
  • Given a candidates table and a results table, write a query that produces a per-candidate report showing their full name and their top 3 states by vote count. Each place column should show 'State (votes)'. If multiple states tie at the same rank, list them alphabetically joined by commas in that column. The ranking itself should be based on distinct vote totals, so tied states share a rank.

“The dense rank part is what gets you.”

View Post

Point72·Machine Learning Engineer·Hiring Manager Screen

May 2026
Screening call for an ML Engineer role on a new agentic AI team at Point72. The whole thing was basically a filter for whether you've actually trained a model from scratch, and they were pretty upfront about not wanting people who've only done LoRA or fine-tuning.
  • Have you trained a large language model or Transformer fully from scratch? Walk me through the entire process, from data pipeline to training infrastructure to evaluation.
  • If your background is mostly fine-tuning or parameter-efficient methods like LoRA or adapters, how do you articulate what you have and haven't done, and what's the actual gap?
  • If you don't have full pretraining experience, what would your ramp-up plan look like once you joined the team?

“This is the core question and they really mean end-to-end.”

View Post

Point72·Software Engineer·Onsite - Behavioral / Leadership

Apr 2026
Behavioral round for a Data Engineer role at Point72. Just one question shared here but it was the kind that takes a bit to think through if you haven't prepped it.
  • Can you walk me through a time you had to work with multiple stakeholders to reach a solution that worked for everyone?

“I had a decent example ready but I rambled a bit trying to cover too many people involved in the story.”

View Post

Point72·Software Engineer·Online Assessment (OA)

Apr 2026
Point72 gave me a 30-minute coding task for a Data Engineer role, basically a stock trading optimizer problem with some extra scaffolding requirements on top of the core algorithm. More structured than I expected for the time limit.
  • Given an array of stock prices, implement a solution to find buy/sell transactions that maximize total profit. Return each transaction as a tuple of (buy_day, buy_price, sell_day, sell_price, profit) and include a report function that prints a summary with total profit. Also write a decorator to validate the input array before processing.

“The core algorithm is the classic greedy approach, buy every dip sell every peak, nothing exotic.”

View Post