Product Manager / Analyst (Analytics & ML)

Location: Remote; must be based in the EU with daily overlap with US Eastern time. The posting also lists Tbilisi and Los Angeles.

Company

Maestra.io is an all-in-one ecommerce marketing platform combining customer data, email, SMS, push, personalization, loyalty, referrals, and product recommendations for mid-market DTC brands.

What you will do

• Own merchant-facing reporting and the semantic layer across attribution, loyalty economics, recommendations, funnels, and segment comparisons.
• Design metrics, slices, aggregates, and partitions; prototype data models in SQL and hand production specifications to engineering.
• Define and audit A/B testing methodology, including group comparability, participant counting, sample sizes, significance, power, and MDE.
• Own the product direction and evaluation of 14 recommendation algorithms, including matrix factorization in Python.
• Monitor recommendation degradation, investigate business performance issues, and evaluate embeddings and LLM re-ranking against baseline models.
• Work directly with engineers, customer success, merchants, and other product managers to turn domain questions into shipped product improvements.

Requirements

• Strong SQL skills, including window functions and performance work over large partitioned datasets.
• Solid A/B testing mathematics and statistics fundamentals, including bias, survivorship, power, MDE, and post-treatment variables.
• Ability to assess embeddings, vector similarity, LLM re-ranking, collaborative filtering, and recommender-system evaluation.
• Direct experience working with non-technical end users and converting their needs into shipped products.
• Clear written communication for asynchronous methodology and product decisions.
• Russian is the working language inside the team; English at B2+ is required.

Nice to have

• Ecommerce or martech experience.
• Ownership of a metric or semantic layer.
• Hands-on exposure to recommender or ranking systems.

Culture & Benefits

• AI-first workflow with extensive use of AI for SQL, Python, ML, and other work, subject to verification.
• Direct collaboration in a founder-funded company of approximately 40 people.
• Horizontal work with product, R&D, customer success, and merchants without multiple management layers.
• Tools include Delta Lake, Trino, Spark, Python, Metabase, Notion, and Slack.
• Product decisions are made quickly, and shipped work reaches merchants within weeks.