Старший DS инженер в команду Авито Путешествия

About the Company

Avito.Travel is a rapidly growing product within the world's largest online classifieds platform. We compete with major travel players and possess a unique advantage: Avito's giant ecosystem, access to user behaviour data beyond travel, and infrastructure that scales solutions to tens of millions of users. Dynamic pricing is one of the business's key bets for 2027 and beyond – ensuring every ruble invested in promotions or discounts works precisely towards its goal: attracting new buyers, reactivating dormant users, and increasing host LTV.

Responsibilities

You will build ML systems for pricing, demand forecasting, and promo allocation – from idea to production. You will design experiments (A/B, causal inference), deploy models into the product, and directly influence business decisions through data.

Example future tasks:

• Per-buyer promos: a model that determines which offer (discount, cashback, bonus nights) to give to which user to maximise conversion while controlling budget.
• Per-item promos: selecting listings to promote in search results – which ads to "highlight" to increase GMV and improve supply-demand matching.
• Uplift models for promo campaigns: distinguishing users who are genuinely moved by a promo from those who would have purchased anyway.
• Price elasticity of demand models: how bookings change when price changes by X% for a given segment / region / season.
• Demand forecasting: short- and medium-term forecasts at the level of regions, dates, and accommodation categories – to optimise promo budget allocation and inventory planning.
• Design and analysis of A/B tests for pricing and promo interventions (accounting for marketplace network effects, switchback designs).
• Causal inference: estimating the effect of promos and price changes when clean A/B testing is not possible (difference-in-differences, synthetic control, instrumental variables).

You will be expected to:

• Design and build ML systems for pricing, promo allocation, and demand forecasting – from problem formulation to production.
• Decompose business problems into ML/optimisation formulations, understanding exactly what we need to optimise and when to use ML versus simpler approaches.
• Design and analyse experiments (A/B, switchback, causal inference), validating models on offline and online metrics.
• Integrate models into the product: real-time/batch pipelines, integration with backend services.
• Ensure ML system maturity: monitoring, alerts, tests, documentation, and experiment reproducibility.
• Influence product and business decisions through data – explaining results to stakeholders and product managers.

Requirements

• 4+ years of experience in Data Science with a focus on pricing, monetisation, promo optimisation, or related problems.
• Proficient in Python and experience building ML pipelines from dataset to production (sklearn, scipy, boosting libraries, PyTorch – depending on the task).
• Experience productising models: Docker, Git, CI/CD, microservice architecture.
• Ability to work with big data: SQL at the level of complex analytical queries, experience with Vertica/Trino or equivalents.
• Understanding marketplace economics: two-sided effects, commission mechanics, and promo mechanics.
• Ability to explain complex models and their results to non-technical audiences – product managers, business stakeholders.
• Self-sufficiency: ability to set your own goals, prioritise, and drive tasks to completion.

Working Conditions

• Opportunity to influence the business and product development.
• Interesting and diverse tasks: Avito analysts identify business growth points, study user behaviour, design frameworks, and set up dashboards.
• High-quality data, powerful infrastructure and tools, any necessary hardware – all ready for productive work.
• A talented team, strong analytical culture, and a community of professionals.
• Transparent bonus system, competitive salary – amount to be discussed during the interview.
• Personal learning budget for books, courses, and conferences.
• Health care: voluntary health insurance with dentistry from day one, on-site therapist and massage therapist at the office.
• Remote work and a great office two minutes from Belorusskaya metro station: panoramic views of the city centre, spaces for focused work and relaxation areas.