Staff ML Engineer (AdTech) | Uzum People

Uzum Market is part of the Uzum ecosystem (Market, Bank, Tezkor, ...). We are building convenient e-commerce for Uzbekistan: we help sellers quickly go online, and buyers find the products they need at fair prices with reliable delivery. We work in a product-oriented way, experiment a lot, make data-driven decisions, and value engineering time.

• The first tech unicorn of Uzbekistan
• Recent investments from major players: Tencent + VR Capital
• The Uzum Market app is consistently in the top-10 in Uzbekistan

The Discovery & Promotion unit includes 3 cross-functional teams (Search, RecSys, AdRev), responsible for all product listings on the marketplace: which candidates should be shown, how to rank them, how to account for organic listings vs. paid promotion.

In this new role, the primary focus will initially be on AdRev (advertising revenue) — one of the key levers for growth and sustainability of any marketplace: it's both a direct contribution to PnL and the main tool that helps new sellers gain traction and find their audience. We are at the beginning of this journey: currently, advertising exists in a basic form in search and catalog, and there is a lot of space ahead:

• Adding CPO (cost per order) as an alternative payment method for sellers;
• Introducing personalization;
• Adding new surfaces (shelves on the product page, feed on the homepage);
• Evolving the ML stack from a CTR model to a system that optimizes money and balances the interests of the user, seller, and platform.

The Staff ML Engineer will report directly to the Head of ML, accelerating and strengthening the current ML teams in search and recommender systems.

The first goals will be related to AdRev. Our goal is for AdRev competency to become native within the ranking of search, catalog, and recommendation surfaces, with this role acting as a horizontal force multiplier: methodology, architecture, solution quality, effect measurement, and rapid deployment to production.

What you will do

• Review the current advertising system: how targeting, logging, models, ranking, and experimentation are set up - identify bottlenecks and the most costly errors.
• Develop CTR models for ad placements: features, architecture, training/calibration, robustness to drift, working with sparse signals.
• Launch the first version of personalization in advertising: user/product embeddings, contextual features, cold start strategies.
• Design data pipelines for advertising ML: collecting and labeling events together with product analysts, improvements in current attribution, data marts, DQ checks, reproducibility.
• Work with multi-objective optimization with constraints: balance organic and advertising content - we definitely don't want to flood everything with ads, degrading user experience and long-term retention.
• Speak the language of money: calculate increment, build business cases, understand unit economics, participate in choosing optimization goals and guardrails.
• Work closely with stakeholders in seller cabinet teams, marketing, and commerce: discuss goals, constraints, success metrics, trade-offs, and expected effects.
• Set the standard: from task formulation and metric selection to production, monitoring, and measurable business impact (including culture of engineering maturity, reviews, design, documentation).
• Together with Search and RecSys, bring changes to production and transfer sustainable ownership: so that teams can develop the system further without a bus factor on a single Staff member.

What we expect

• 7+ years of experience in DS/ML Engineering: built models and systems that actually live in production, understand the constraints of latency/cost/reliability.
• Mandatory experience in marketplace/e-commerce or AdTech, specifically in tasks where money is optimized (revenue/margin/ROI/GMV-value), not just "model quality."
• Strong Python + SQL, engineering culture: clean code, reviews, system design, documentation.
• Ability to work in "messy reality" conditions: curiosity to dig into data, find inconsistencies, gather correct metric definitions.
• Understanding of experiments and causality: metrics, design, interpretation of results, guardrails.
• Communication at the Staff level: able to explain complex things in simple terms and negotiate goals/constraints between teams and business.

It will be a plus if you have 3 out of 5

• Experience building advertising systems and auctions.
• Experience in search/recommendations (general ranking principles, retrieval → ranking, quality metrics).
• Launched more than 20+ A/B tests, worked with complex metric trees.
• Experience with multi-objective optimization and strict constraints.
• Experience combating fraud/spam/manipulation in advertising signals.

Why it's interesting:

• Modern stack - Python, NumPy, Pandas, scikit-learn, CatBoost, PyTorch, Hugging Face, ONNX, Elasticsearch, ClickHouse, PostgreSQL, Airflow, Spark, Docker, Kubernetes, Kafka, GrowthBook.
• We provide a wide area of responsibility and the opportunity to influence architectural and product decisions. We are also happy to discuss your initiatives and implement them.
• Unique culture – we have preserved the startup spirit while already establishing mature processes.
• We set measurable goals as a team and don't just complete them "for the sake of it," but measure effectiveness and the overall impact of our work on the business.
• Work in teams of strong specialists, where depth of expertise and engineering thinking are valued.
• Teams listen and hear each other, acting as partners, not just executors.

What we offer:

• Remote work from anywhere in the world or a cozy office in Tashkent.
• You can grow in the engineering or management track, and we have a regular performance review system in place.
• We pay at the level of top companies in the Russian market.
• Learning and development — we support both within the company and outside (meetups, conferences, professional training, publications). We also help develop your personal brand.
• The base is a community of professionals with a desire to do great things. A nice bonus — health insurance tied to your location, training, and other perks.