Software Development Engineer - Machine Learning, Alexa Smart Home Machine Learning Infrastructure at Amazon
Get hired in a next generation AI company right now.
"Craiglist for jobs in AI" - Tony Rose
As seen in
🖤🤍 If you like MoAIJobs, give us a shoutout on 𝕏
Filter by:
Amazon
Software Development Engineer - Machine Learning, Alexa Smart Home Machine Learning Infrastructure
US, WA, Seattle
machine learning
LLM
generative ai
We are looking for a Machine Learning Engineer that will help us evolve Alexa into the AI assistant of the future by helping build highly scalable Generative AI systems and infrastructure. The Smart Home Machine Learning team is focused on making Alexa the user interface for the home. From the simplest voice commands (turn on the kitchen light or turn down the heat) to more complicated and natural use cases that use the power of AI and LLM, such as “Turn on the new light in my office” or “Turn off all the lights, except the living room if the TV is on”. We are evolving Alexa into an intelligent, indispensable companion that automates daily routines ("Turn on my outdoor lights at sunset every weekday”), simplifies interaction with appliances and electronics, and alerts when something unusual is detected. You will be part of a team delivering features that are highly anticipated by media and well received by our customers. And, you will have the satisfaction of working on a product your friends and family can relate to, and want to use every day.
The key unique challenge in the Smart Home domain is that the interpretation of all utterances is dependent on the device context of the home. This is the collection of all smart home and echo devices that a customer owns, along with the names and grouping of those devices, as configured by the customer. This means that for the same utterance ("Turn on the lights"), a different device context results in different correct actions, such as turning on a single light, all lights, or a smart plug. Solving the problem takes on complexity in these two dimensions, which we call utterance complexity and customer complexity.
This year we are launching an updated LLM based Alexa SH that will expand on the existing use cases already supported by our Classic Speechlet. There are two main themes we are launching this year: Our first theme is conversational VUI, where the LLM will greatly extend our existing NLU/entity resolution capabilities to easily handle voice requests we don’t support today. Some examples include “Alexa, lock the door and turn off the lights” in which the customer concatenates two unrelated control directives, “Alexa, turn on the kitchen lights except the sink one” in which the customer targets a complex collection of devices, or “Alexa, its dark in here” in which the customer’s intent is ambiguous and we expect the LLM to understand which smart devices relate to the request (i.e. lights, blinds) and reason about how to control them to satisfy customer’s request. Our second theme is automation VUI by which we want to allow customers to create routines thru voice which support all of the actions available through conversational VUI. Examples include “Alexa, lock the door when I leave” or “Alexa, turn off the lights and TV in 5 minutes.”
Key job responsibilities
As an SDE on the team, you will lead development design patterns, APIs, and high-scale services for machine learning that make the Smart Home intelligent, including Alexa Hunches. Your work will span Alexa skills, voice user interfaces, cloud services, and a rapidly-growing ecosystem of IoT devices. You will have the satisfaction of working on a product your friends and family can relate to, and want to use every day. Like the world of smart phones less than 10 years ago, this is a rare opportunity to have a giant impact on the way people live.
A day in the life
As a Machine Learning Engineer on the Smart Home Machine Learning team, you will combine engineering best practices with your knowledge and experience with Machine Learning to lead the building of cutting-edge infrastructure components and platforms that are highly scalable, extensible and robust to enable exponential growth and adoption of Machine Learning applications within the Organization. You will work very closely with science and engineering leaders to help innovate and define the vision of products and features you own.
About the team
The Smart Home Machine Learning Infrastructure team aims to build cutting-edge infrastructure components and platforms that are highly scalable, extensible and robust to enable exponential growth and adoption of Machine Learning applications within the Organization. The customers include customers of Alexa and Scientists and Engineers. The team thrives on an inclusive culture of diverse ideas and works on spending dedicated time to facilitate learning and exchange of ideas.
The key unique challenge in the Smart Home domain is that the interpretation of all utterances is dependent on the device context of the home. This is the collection of all smart home and echo devices that a customer owns, along with the names and grouping of those devices, as configured by the customer. This means that for the same utterance ("Turn on the lights"), a different device context results in different correct actions, such as turning on a single light, all lights, or a smart plug. Solving the problem takes on complexity in these two dimensions, which we call utterance complexity and customer complexity.
This year we are launching an updated LLM based Alexa SH that will expand on the existing use cases already supported by our Classic Speechlet. There are two main themes we are launching this year: Our first theme is conversational VUI, where the LLM will greatly extend our existing NLU/entity resolution capabilities to easily handle voice requests we don’t support today. Some examples include “Alexa, lock the door and turn off the lights” in which the customer concatenates two unrelated control directives, “Alexa, turn on the kitchen lights except the sink one” in which the customer targets a complex collection of devices, or “Alexa, its dark in here” in which the customer’s intent is ambiguous and we expect the LLM to understand which smart devices relate to the request (i.e. lights, blinds) and reason about how to control them to satisfy customer’s request. Our second theme is automation VUI by which we want to allow customers to create routines thru voice which support all of the actions available through conversational VUI. Examples include “Alexa, lock the door when I leave” or “Alexa, turn off the lights and TV in 5 minutes.”
Key job responsibilities
As an SDE on the team, you will lead development design patterns, APIs, and high-scale services for machine learning that make the Smart Home intelligent, including Alexa Hunches. Your work will span Alexa skills, voice user interfaces, cloud services, and a rapidly-growing ecosystem of IoT devices. You will have the satisfaction of working on a product your friends and family can relate to, and want to use every day. Like the world of smart phones less than 10 years ago, this is a rare opportunity to have a giant impact on the way people live.
A day in the life
As a Machine Learning Engineer on the Smart Home Machine Learning team, you will combine engineering best practices with your knowledge and experience with Machine Learning to lead the building of cutting-edge infrastructure components and platforms that are highly scalable, extensible and robust to enable exponential growth and adoption of Machine Learning applications within the Organization. You will work very closely with science and engineering leaders to help innovate and define the vision of products and features you own.
About the team
The Smart Home Machine Learning Infrastructure team aims to build cutting-edge infrastructure components and platforms that are highly scalable, extensible and robust to enable exponential growth and adoption of Machine Learning applications within the Organization. The customers include customers of Alexa and Scientists and Engineers. The team thrives on an inclusive culture of diverse ideas and works on spending dedicated time to facilitate learning and exchange of ideas.