Senior Applied Scientist at Amazon

Short Description:

The Senior Applied Scientist role at Amazon's Shipping Tech and Services team in multiple Indian locations requires 3+ years of ML model building expertise, along with a PhD or Master's and 6+ years of applied research experience. Candidates need proficiency in C/C++, Python, Java, or Perl, deep learning methods, and neural networks. Familiarity with R, scikit-learn, Spark MLLib, and distributed systems like Hadoop is preferred for this impactful role crafting cutting-edge ML solutions for efficient package movements and predictive delivery enhancements.

---

 

Senior Applied Scientist

Job ID: 2479698 | ADCI - Haryana

DESCRIPTION

Our customers have unwavering confidence in our ability to deliver packages promptly and precisely. An intelligently designed network effortlessly scales to manage millions of package movements daily. It incorporates monitoring mechanisms that can identify and preempt failures (e.g., predicting network congestion, operational disruptions), and take proactive corrective measures. When setbacks do occur, it has built-in redundancies to reduce the impact (e.g., finding alternate routes or service providers capable of handling increased loads) and avoids dependence on single points of failure (service providers, nodes, or arcs). Moreover, it is cost-effective, ensuring customers benefit from an efficiently organized network.

The Shipping Tech and Services (STS) team is seeking Senior Applied Scientists to enhance our ability to strategize and carry out package movements. As a Senior Applied Scientist at STS, you will lead a pioneering team of scientists who employ Machine Learning to craft cutting-edge solutions for Amazon's customers. You will analyze and model terabytes of data to address real-world challenges. You will be responsible for end-to-end business issues/metrics and will have a direct impact on the company's profitability. Your models will forecast shipping costs on Day 1 and enhance the quality of upstream data (e.g., by correcting weights and dimensions in the product catalog). By utilizing information from within the transportation network (e.g., network load and the speed of movements derived from package scan events) and external data (e.g., weather signals), you will develop models to predict delivery delays for every package. These models will enhance the customer experience by triggering early corrective actions and generating proactive customer notifications.

Your role will require you to embody the principles of "Think Big" and "Invent and Simplify" by refining and translating transportation-related business challenges into one or more Machine Learning problems. Your model options will encompass a range of choices, including tree ensembles, deep learning architectures like LSTMs and transformers, large language models, and Q-learning models. You will employ advanced model interpretation techniques such as LIME and SHAP to ensure transparency and establish trust with customers. You will lead high-impact business initiatives and provide guidance to other scientists and engineers in applying innovative ML techniques. You will create a set of reusable modeling solutions that leverage techniques like transfer learning and few-shot learning, ensuring scalability across multiple regions (e.g., North America, Europe, Asia) and types of package movements (e.g., small parcels and truck movements). Your models will be of production quality, directly influencing core production services with the latest ML advancements.

You will work as a part of a diverse data science and engineering team, which includes other Applied Scientists, Software Development Engineers, and Business Intelligence Engineers. You will be actively involved in the Amazon ML community, contributing by authoring scientific papers and submitting them to Machine Learning conferences. You will serve as a mentor for Applied Scientists and Software Development Engineers with a strong interest in ML. Additionally, you may be called upon to provide ML consultation for other teams facing unique challenges.

We are considering candidates for positions in the following locations:

• Bangalore, KA, IND
• Gurugram, HR, IND
• Hyderabad, TS, IND

BASIC QUALIFICATIONS

• 3+ years of experience in building machine learning models for business applications
• PhD, or Master's degree and 6+ years of applied research experience
• Proficiency in programming languages such as C/C++, Python, Java, or Perl
• Proficiency in programming languages such as Java, C++, Python, or related languages
• Proficiency in neural deep learning methods and machine learning

PREFERRED QUALIFICATIONS

• Experience with modeling tools such as R, scikit-learn, Spark MLLib, MxNet, Tensorflow, numpy, scipy, etc.
• Experience with large-scale distributed systems, such as Hadoop, Spark, etc.

 Please click here to apply.