Engineering Full-Time

Senior Machine Learning Engineer (US)

Petra Security • San Francisco, CA

Company

Petra Security

Annual salary

$250k – $300k/yr

Location

San Francisco, CA

Listed

54d ago

Experience:: 3 to 8 years
Workplace:: 5 days in-office in San Francisco
Equity:: Competitive equity

On-site in San Francisco, CA.

Send us your LinkedIn and CV. If your experience fits, we'll introduce you to the recruiter filling this role.

Apply for a referral →

The recruiter emails you before anything happens and may suggest other jobs that suit you better.

What you'll do

Ownership is everything here. You'll be responsible for our core ML detection platform, and the systems you build are what separate Petra from every other player in this space. Petra is a premium product and it's everyone's job to make sure it stays that way. That means you won't just train models, you'll talk to customers, understand the attacks they're facing, and think creatively about how to find the next signal that widens the gap between us and second place. Concretely, you will:
Build and evolve our detection models. Working closely with our CTO and Head of Machine Learning, you'll design and build the ML systems that identify account compromise in real time.
Take models from research to production. You don't just care about model performance in a notebook. You're energized by measuring how your models behave at scale in a live system with real attackers.
Architect for scale. We're growing fast. Nothing excites you more than architecting and implementing an elegant solution that'll handle 10-100x the scale in 6 months.
Raise the technical bar. As a senior voice on a lean team, your instincts and opinions shape how we build. You'll establish patterns, push for correctness, and raise the quality of everything around you. Our stack is Node, TypeScript, React, and Python... For now.

What you need

4+ years of machine learning engineering experience, with meaningful time shipping models in production systems at scale.
**Strong quantitative fundamentals: **probability & statistics, linear algebra, anomaly detection, behavioral modeling, NLP.
**Experience training & deploying models in production: **You know how to pick the right model for the job. Just as importantly, you know the limitations of each modeling approach you consider.
Strong engineering fundamentals, you write clean, production-grade code and are comfortable owning a system end to end, not just the modeling layer.
Experience working with real-time or streaming data pipelines, feature stores, high-performance databases, and distributed systems. Bonus if you've worked in high-stakes domains like quantitative finance or fraud detection.
High agency and high output. You identify the problem, propose the solution, and execute. You're constantly looking for ways to expand your scope and circle of competence.
Sharp and scrappy. You learn fast: you ask the right questions and figure it out. You ship fast: you find the 90/10 and don't let perfect be the enemy of done.
Comfortable in a small-team, high-autonomy environment. The scope of your work is rarely handed to you, and you wouldn't have it any other way.
Proud of your craft. You care about building things that work well and hold up over time.

The Team

We're small and high-caliber. Our founders are ex-Abnormal Security, Palo Alto Networks, and Bridgewater. Other early team members include:

A Sequoia-backed founder
An engineering leader from Anduril
Quants from Citadel and Point72

You'll work directly with people who are exceptional at what they do.

Frequently Asked Questions

How do I apply for the Senior Machine Learning Engineer (US) role at Petra Security? +

Use the Apply for a referral button on this page to send us your LinkedIn and CV. If your experience fits, we'll introduce you to the recruiter filling this role. They'll email you to check you're interested, then put you forward for this job or others that suit you better. It's free.

What does this Petra Security role pay? +

The listing gives $250k – $300k/yr.

Is this role remote? +

The listing gives the location as San Francisco, CA. 5 days in-office in San Francisco.

Interview Prep

Sample questions for a Machine Learning Engineer role, written in-house to help you prepare.

How do you choose between a simpler interpretable model and a more complex model that performs marginally better?

The decision depends on how the model's output is used downstream. If a human needs to act on and justify individual predictions, like in lending or healthcare, interpretability often outweighs a small accuracy gain. If the model feeds an automated system where explainability isn't a hard requirement, I'll take the complexity if the performance gain is meaningful and validated, not just noise.

Explain how you would set up a proper train, validation, and test split for a time series problem.

Random splitting leaks future information into training for time series, so I split chronologically instead, training on the earliest period, validating on the next, and testing on the most recent. Any cross-validation also needs to respect time order, using something like rolling-origin validation, rather than standard k-fold, which would otherwise let the model see the future during validation.

What statistical test would you use to determine if an A/B test result is significant, and what assumptions does it rely on?

For a conversion-rate comparison, a two-proportion z-test or chi-squared test is standard, and it assumes independent observations and a large enough sample size for the normal approximation to hold. If the sample is small or the metric isn't binary, I'd switch to a more appropriate test, since applying the wrong test's assumptions is a common source of false confidence in A/B results.

How do you approach hyperparameter tuning efficiently when the model is expensive to train?

I use a coarse random search first to identify the promising region of the hyperparameter space, since grid search wastes compute on combinations unlikely to matter. From there, Bayesian optimization narrows in on the best configuration with far fewer expensive training runs than an exhaustive search would require, which matters a lot when each run is costly.

See all 10 questions for this role →