Tech Lead for Machine Learning Quality Group in e-Commerce Content System

Tech Lead for Machine Learning Quality Group in e-Commerce Content System

Search technologies are the DNA of the Search business unit. Already, one in five queries in Search is for product selection. This scenario accounts for 40% of the profit. We are working on a tool that searches information across all possible online stores (over 60,000). According to plans, a convenient AI consultant will be integrated into the tool: it will help compare products by characteristics or decide where it's better to buy.

We are looking for an ML Tech Lead who will head the streams of product matching and SKU creation in the Product Search team.

What tasks await you

Leading the development team The vacancy implies a player-coach role, within which approximately half of the working time will be spent on working with the team (code reviews, discussing solutions, planning tasks, etc.), and the second half on independently solving the most critical and complex tasks. The role also implies interaction with customers, receiving tasks from them, and presenting the results of the work.

Setting up the product matching process Our team prepares data that allows comparing prices for products across the entire e-commerce catalog of Russia. This task is called matching, and to solve it, one must match two product cards from different sellers and determine if it's the same product or different ones. The complexity of the task lies in the fact that to solve it, all data about the product (images, description, attributes) must be taken into account, and also stability of trained models must be achieved so that they work correctly with both popular devices like iPhone 16 Pro Max and niche ones like plumbing pipes.

Creating SKU cards Automatic creation of SKU cards is a fundamentally complex task. You will need to create new SKUs based on offers from various e-commerce platforms, in whose cards information about the product is indicated in as much detail as possible and beautiful, relevant pictures are placed. The difficulty of the task lies, firstly, in not creating a new SKU when a suitable one already exists in the database, and secondly, in filling the SKU card with combined (sometimes even contradictory) information from several offers.

We expect that you

• Have industrial experience with NLP (including LLM)
• Have managed a team of several developers
• Have optimized ML models and deployed them to production
• Have built automated machine learning pipelines
• Can formulate tasks in ML terms, have solved ML tasks in practice
• Understand how to measure results, are familiar with different algorithms and can choose the appropriate one
• Know Python and SQL well, write clean code

Will be a plus

• Have dealt with CV
• Have set up matching in a large e-commerce
• Have been involved in commercial development
• Know C++ or are ready to switch to it