Data Scientist

About the company:

At PrettyDamnQuick, we help ecommerce brands win- one checkout at a time, by providing an Amazon-like experience..

We’re solving a $270B problem where 30–50% of customers abandon checkout. Our AI-driven Autonomous Checkout learns, segments, and optimizes in real time to boost conversion, AOV, and loyalty from click to delivery.

Trusted by 250+ fast-growing brands and backed by top-tier investors, PDQ is where personalization becomes profit- and the future of ecommerce gets built.

If you’re driven by impact and obsessed with building what’s next in ecommerce, welcome home.

About The Role

We are looking for a Data Scientist to join our Data team in Tel Aviv and own the science behind PDQ’s products.

PDQ sees tens of millions of shoppers across 200+ merchant checkouts, and around half of the shoppers active in a given month have already been seen at another shop we serve. You turn that view into better decisions: who this shopper is across the network, what they are worth, what to show them, and which coupon or shipping option to offer at checkout. At most companies these sit in different teams. Here they are one problem, and yours.

This is a hands-on role. You choose the modelling approach, build it, ship it, and carry it in production. You join a small, full-stack data team as its first data scientist, so you set the bar. Individual contributor with no direct reports at the start, with room to grow the team as the work scales.

What You’ll Own

• Shopper identity across the network (PDQ ID). How we recognize the same shopper across merchant checkouts, and how confident we are when we do.
• Predicted customer value. What a shopper is worth across the network, not to a single merchant.
• Recommendation and personalization. What to show and what to upsell, informed by behaviour at other shops.
• Autonomous Checkout decisioning. Which coupon and which shipping option to offer per checkout, and the learning loop behind it.
• The modelling foundation. Choose the approach (classic ML, recommenders, causal, bandits) and defend it with evidence.
• Models that hold up across 200+ merchants with different catalogues, margins and shoppers.

Requirements:

• 5+ years of hands-on experience as a data scientist, machine learning engineer, or similar role, including at least two roles where you owned models in production end to end.
• Depth in at least one of: customer value modelling (LTV), recommendation systems, causal inference and uplift, bandits and reinforcement learning, or entity resolution. You do not need all of them.
• You lead a project end to end: scope it, make the trade-offs, ship it, and own it in production.
• Experimentation fundamentals. You design tests that answer the question, and you know when an offline estimate can be trusted.
• Strong SQL and Python. You can code without AI, and you use AI development tools well.
• You can explain a modelling decision to people who do not model, in business language.
• Ability to work in our Tel Aviv office, four days a week.
• Full professional proficiency in English (written and spoken).

Advantages

• Experience at an e-commerce company, or at a company that sells software or services to e-commerce.
• Consumer psychology and business thinking, with curiosity about how shoppers decide.
• Entity resolution, identity graphs, or probabilistic matching at scale.
• Bandits, off-policy evaluation, or reward attribution in a live system.
• Embeddings or LLM representations used inside a decisioning or personalization system.
• Modelling under privacy or consent constraints.
• Models that serve many customers at once, where one system has to hold up across very different businesses.
• MSc or PhD in a quantitative field.
• knowledge in Snowflake