You will take on technical leadership and end-to-end ownership for our Pricing/Revenue-ML topics—with a clear focus on measurable impact. You will work closely with Product and Engineering, define measurability/experiments, and ensure that our models not only “look good” but also perform reliably in practice.
Important: No disciplinary personnel responsibility. You lead through expertise, standards, and ownership.
End-to-End Ownership: You are responsible for the entire lifecycle of pricing and revenue topics—from hypothesis to implementation to measurable evaluation. Your focus: Clear business uplift.
Smart Modeling: You develop and optimize forecasting and pricing models. You pragmatically decide which method gets us to the goal fastest and most stably.
Signal Expertise: You manage time series, demand signals, and heterogeneous data sources. You ensure that features and labels are defined absolutely clean and “leakage-proof.”
Experimentation Framework: You build a robust measurement system (holdouts, A/B tests, guardrails) and define crystal-clear criteria for rollout decisions.
Engineering-Grade ML: You establish standards for backtesting, reproducibility, and versioning. For us, it's: Engineering quality instead of notebook-only.
MLOps Best Practices: You bring MLOps best practices and drive continuous improvement in our ML workflow.
Reliable Operations: You ensure operations through smart monitoring, drift detection, and pragmatic retraining mechanisms.
Automation & Scale: You automate high-leverage processes (backtests, monitoring checks) to massively increase throughput and quality.
Data Foundation: Where it makes sense, you design data models directly in the warehouse (Snowflake/dbt) as a basis for reliable metrics and features.
Full Transparency: You standardize dashboards (e.g., Metabase) for our business KPIs and ensure the data quality is beyond reproach.
Stakeholder Sparring: You prioritize requirements together with Product & Revenue and translate them into ML solutions. Your motto: Impact over output.
Deep Experience: You have 6+ years of experience in applied ML engineering or data science – ideally directly in a product or business context.
Proven Impact: You have already achieved demonstrable success in the areas of pricing, revenue, forecasting, or similar “money systems.”
Evaluation Pro: You think offline vs. online, immediately recognize bias/leakage, and master the fundamentals of robust metrics and guardrails.
Tech Stack: Your Python and SQL skills are production-level (testable, versioned, reproducible).
Startup DNA: You love the 80/20 principle, work extremely pragmatically, and want full ownership for your topics.
Language Skills: You communicate fluently and confidently in English.
Hands-On MLOps & Cloud: Bonus if you have hands-on skills to work with MLOps tooling and cloud infrastructure, e.g. AWS
Domain Knowledge: Experience in revenue management or dynamic pricing (e.g., travel, mobility, eCommerce).
Demand Understanding: You know how seasonality, events, and lead times affect pricing.
Modern Toolchain: You are proficient in analytics engineering (dbt, Snowflake, Metabase) and know how to build a clean data foundation.
Apply and improve machine learning models in active drug discovery programs, validating predictions, guiding experimental teams on model quality and uncertainty, and connecting research with practical discovery needs.
Own the strategy and systems that score, select, and provision large-scale driving datasets for model training, balancing data value and quality with storage costs as the corpus grows.
Leads technical direction for ads conversion machine learning, building low-latency, large-scale prediction models and using text, visual, and user signals to improve ad products.
Define and build the LLM platform and AI systems powering clinical products, from agentic workflows and retrieval to evaluation and production infrastructure, while setting technical direction and mentoring engineers.
Develop and evaluate machine-learning methods that help autonomous-driving models transfer reliably across vehicle platforms, diagnosing real-world failures and improving production performance.
