Denis Platon Email & Phone Number

- Collaborated with game designers and product managers to define and track key performance indicators (KPIs) for various in-game features and assessed the necessity of the implementation of some features- Conducted the analysis of player behavior data to identify gameplay patterns and user preferences, leading to game design decisions that increased overall player retention by 2.5%- Designed and executed A/B tests to evaluate the impact of ad placements and formats, gameplay innovations, new level sequences, different number of boosters per level, leading to an 8% ARPU growth by 14th day.- Developed a player segmentation model based on behavioral analysis, that predicted the segment and adjusted the user experience accordingly

- Built a predictive maintenance model for ESP failures, resulting in a 20% reduction in downtime between pump failure and replacement.- Established the connection between the product development team and users directly, enabling more effective communication, reducing bias- Presented the results of the work to C-level management, demonstrating exactly how the interaction between the Run and Change teams had changed.- Calculated the project efficiency and its main KPIs together with the economic department, which were subsequently demonstrated to the company’s management- Hypothesised the retrofitting of pump sensors for more accurate model predictions, conducted MVP and proved the effectiveness, resulting in a 10% improvement in the model quality and an additional 5% reduction in the time between repairs

Perm, Russia

- Created an MVP of a predictive maintenance model for predicting failures of high production ESPs, generating the profit of approximately 30% of the company's total annual revenue- Built Python scripts reducing the time for uploading data and creating a model by 40%-Changed the data exchange scheme from "requesting data via email in Excel templates" to SQL queries directly from databases, removing the human factor from the data exchange and reducing data collection time by 70%.

Perm, Russia

- Gathered production data from different sources, built a mathematical model of the oil and gas gathering system between 1000 wells, performed debottlenecking calculations and made recommendations that helped reduce pressure in the system by 1.5 bara, allowing 1% more oil to be produced within 6 months- Built a mathematical model of an offshore oil field with 10 wells and optimized gas injection in a gas lift well, allowing 10% less gas to be injected- Developed and implemented Python scripts to automate routine operations in the Integrated Asset Model, resulting in a significant reduction of time spent on manual tasks by 30%

Master'S Degree, Petroleum Engineering

Thesis - Prediction of ESP failure with machine learning models

Master'S Degree, Design, Construction And Operation Of Oil And Gas Pipelines And Storage Facilities, Gpa: 4.8/5