Мы используем технические cookie и статистику посещений: Яндекс Метрику и собственный обезличенный журнал Zorky (cookie zk_aid). Не собираем содержимое форм, текст поиска и полный IP; данные не продаём. Отказаться можно кнопкой ниже, а также через Global Privacy Control или Do Not Track в браузере. Подробнее о cookie и статистике.
Zorky CRMZorky CRM
EN|RU
@termdocs
← Все вакансии

Software Engineer, ML (Time Series Forecasting)

hybridBrooklyn, USСкор 57.5/1002д назад
Аналитика рынка
📊 AI / ML / DS: зарплаты и спрос на рынке
Стек
awscloudfinopspythonsql
Показать компанию и ссылку
Бесплатно. Вход по почте.
Откликнуться
Загрузите резюме — мы свяжем вас с работодателем напрямую через нашу базу.
Отправить резюме →
Описание
About the Role Forecasting sits underneath much of North: projected spend, budgets, and the commitment purchases we make on customers' behalf all depend on knowing where a customer's usage is headed. When a forecast is off, customers feel it, either in missed savings or in commitments that go unused. You'll join the Services team and own the models behind North's internal forecasting service, working with your teammates on the service that delivers forecasts across North's features. You'll shape how the service evolves, from how models are trained and evaluated to how forecast quality is monitored in production. You'll also work closely with data scientists and the teams who rely on forecasts, so results are clear to the people making decisions from them. We're hiring for this role without a fixed level. We'll level you based on experience and on how you work, using our engineering ladder. What You'll Do Build and improve the time series models behind projected spend, budgets, and commitment recommendations, across customers with very different sizes and usage patterns. Partner with your Services teammates to take models into production as a reliable shared service with clear interfaces. Shape the forecasting service's architecture, including how models are trained, versioned, evaluated, and served. Improve how we evaluate, backtest, and monitor forecast accuracy over time. Build proactive detection and alerting for unusual forecasts and model drift, so issu
Полное описание, компания и ссылка на вакансию — после входа. Войти
Контакты работодателя (email/phone/telegram) скрыты из публичного превью — отправьте резюме, чтобы мы связали вас напрямую.
Срочный вопрос? Напишите @termdocs