Скидка на вакансию
Скидка на первую вакансию
Акция до 22 декабря
👀 Бэкендеры, зовем на Публичное собеседование на Хабр Карьере! Для участия нужно → оставить заявку
Amina Kaltayeva (ak424), Казахстан, АлматыAmina Kaltayeva (ak424), Казахстан, Алматы

Amina Kaltayeva

Эксперт
Ученый по даннымАналитик по даннымСтарший (Senior)
От 3500 €Рассмотрю предложения

Контакты

Войти
Опыт работы: 7 лет и 10 месяцев
Регистрация: 25.01.2022
Последний визит: 2 дня назад
Гражданство:
Местоположение: Казахстан, Алматы
Дополнительно: готова к удаленной работе
Знание языков:

Обо мне

Data Scientist with more than 5 years of experience, highly skilled in descriptive and predictive analytics, used to working with different types of data (quantitative/qualitative, textual, geo, time-series) of varying quality and translating them into a coherent story.

Key hard skills: python (including pandas/numpy/sklearn/matplotlib/pytorch) / SQL/ PowerBI 

Key soft skills: as a former consultant I am a quick-learner, can think critically, conduct contextual interviews, work independently with no supervision and present my ideas in concise manner with good-looking delivarables 

Навыки

Анализ данных
Машинное обучение
Python
SQL
Управление проектами
Управление рисками
Нейронные сети
Обработка естественного языка
Исследование пользователя

Эксперт

4консультации
5
4 оценки

Работаю с квалификациями

СтажёрМладшийСредний

Работаю с запросами

Экспертность

Я data scientist с более чем 5 годами опыта работы. Прошла путь от выпускника экономического факультета, внешнего консультанта по управлению рискам\бизнес-процессами до data scientist-a. Все дополнительные навыки (сертификат по управлению рисками - FRM, умение программировать на Python-е, практическое применение машинного обучения) нарабатывала, не прерывая рабочий опыт.  

Могу помочь выстроить план карьерного развития, план получения необходимых навыков для продвижения по карьерной лестнице/ смены профессии так, чтобы максимально задействовать текущий опыт и навыки и максимально не отрываться от рабочего процесса. 

Также как опытный data scientist могу помочь с рабочими задачами касательно машинного обучения. 

Стоимость
3000 ₽/ час
Первая встреча бесплатно

Участие в профсообществах

Опыт работы

  • Ученый по данным (Старший)Machine learning developer
    Июнь 2022 — Февраль 2024 (1 год и 9 месяцев)
    • Full-stack predictive modeling: collecting data internal/external data, data processing, primary data analysis on trends and patterns, feature engineering, ML algorithm selection/tuning/validation, testing, production-grade code writing. Models delivered into production: three scoring models, including main one, for all online clients. Increased Gini of scoring models on 1-3% on average.
    • Uplift modeling. Built from scratch the soft/hard collection models (for optimization work of collectors and selecting clients who are most likely to pay after different collection actions).
    • Developed clients profiles (geo, mobile activity, payments) with ML and pattern analysis models for prediction and monitoring of main product metrics and anomalies. 
    • Introduced reject inference methods to scoring models building.Increased Gini of scoring models on 5-7% on average.
    • Built automated machine learning pipeline for routine tasks that automated: raw dataset uploading, cleaning, primary data analytics building, binning of features, base model creation. Sped up ML model development process by about 3-4 days.
    • Designed and conducted client tests to collect data for predictive modeling. For example, designed data collection of the post-overdue client behavior. Used collected data for building ML models. 
    • Created dashboards to evaluate and monitor ML models' performance and stability of population: developed different data pipelines and organized in executable python script/BI reports/ DB table. For example, highlighted external data provider with 35% non-response rate.
  • Универсальный коммерческий банк Казахстана
    АлматыБолее 5000 сотрудников
    Ученый по данным (Средний)Middle Data Scientist
    Ноябрь 2021 — Июнь 2022 (8 месяцев)
    • Full-stack scoring model development. Provided comprehensive analysis solutions to build scoring, PD, LGD, EAD models using data from internal and external sources and applied advanced analytical methods to assess factors impacting creditworthiness of SME clients. Applied logistic regression/ tree-based models/ NN methods.
    • Growth analytics. Built customer journey map to evaluate conversion rate at each step from loan request to loan issue. That exercise highlighted the need to introduce client pre-scoring (that is, target potential clients before they apply).
    • Developed monitoring dashboard to track ML models' performance, stability of population and key product metrics: dashboard included factual information about product performance and dynamic forecast of achievement year-end targets.
  • Казахстанско-Британский технический университет
    Алматы
    Senior Lecturer
    Август 2019 — Июнь 2021 (1 год и 11 месяцев)
    • Employed advanced knowledge of risk management and data science to teach challenging high-level courses (Microeconomics and Quantitative finance) to undergraduate students. Courses included econometrics and time-series models.
    • Increased final exam passing rate by about 20%.
    • Delivered undergraduate courses via classroom, online and hybrid methodologies to satisfy course curricula and meet student needs.
  • Международная сеть компаний, предлагающих услуги в области консалтинга и аудита
    АлматыОт 1000 до 5000 сотрудников
    Senior Consultant, Risk Assurance Services
    Сентябрь 2017 — Октябрь 2021 (4 года и 2 месяца)

    Main projects and responsibilities:

    • End-to-end analytics system and implementation of unit economics. For large retail company re-built managerial reporting by introducing unit economics. That allowed developed the end-to-end analytics, prioritized expenses and cut next-year budget by 11%.
    • Customer segmentation model. For large retail bank created segmentation model in Python to identify client profiles and marketing channels with highest CAC. Model is also used for additional product recommendations and new clients classifications. 
    • Business processes re-engineering. For different companies reviewed business processes, created user-stories, identifying areas that required improvement and conducted process re-engineering. Processes included IT, procurement, sales, HR, budgeting. Applied data visualization and gap analysis.
    • Quantitative risk assessment framework. For telecom company developed quantitative risk assessment model that helped to formulated knowledgeable forecasts for business and industry based on available financial information. Applied simulation techniques.
    • Conducted seminars to teach clients how to assess and control risks at front line.
    • Orchestrated multiple projects within strict timeframes and budget constraints by solving complex problems and working closely with senior leaders. Coached, developed and motivated team members, providing coaching and mentoring to junior colleagues.
  • Один из мировых лидеров отрасли FMCG
    АлматыБолее 5000 сотрудников
    Intern, Demand planning, Supply chain
    Июнь 2015 — Сентябрь 2016 (1 год и 4 месяца)

    - Managed of short-/long-run demand planning for two countries out of CEAB unit (contained 13 countries);
    - Conducted post-launch reviews of marketing activities on impact on sales; 
    - Was responsible for sales-out data management
    Key achievements: (1) delivered a short/long-term demand forecast statistical model that allowed to rise forecast accuracy from 80% to 85; (2) designed and implemented an inputs collection worksheet that optimized period-close cooperation of several departments and saved around 4 working hours per period, (3) created a system of products’ unique codes in order to collect sales out data for further analysis

Высшее образование

  • LSE-London

    London school of economics and political science
    Лондон50 выпускников
    Accounting and Finance
    Сентябрь 2014 — Сентябрь 2017 (3 года)

    - BSc Economics and Finance (First Class)

    - Main subjects: Mathematical statistics (including probability theory), Econometrics, Qualitative Finance (models for financial data), Microeconomics

  • КБТУ-Алматы

    Казахстанско-Британский технический университет
    Международная школа экономики и социальных наук; МШЭ
    Август 2013 — Июнь 2017 (3 года и 10 месяцев)

    - Бакалавр Экономики и Финансов

Дополнительное образование