Amazon Technologies, Inc.

Scott Gerard Carmack

Narasimha Rao Lakkakula

Nima Sharifi Mehr

Sachin Purushottam Joglekar

Ashish Kumar Agrawal

Amit Bhosle

Dhanvi Harsha Kapila

Shilpi Gupta

Mehul Jain

Daniele Bellucci

Jon Arron McClintock

Samuel Scott Gigliotti

Ryan Justin Snodgrass

William Frederick Kruse

Thibault Candebat

Joel C Hegg

Alun Jones

Sunu Aby Mathew

Ian T R MacLeod

Slawomir Kosowski

Shailendra Batra

In response to a process being triggered, at least in part by receipt of information regarding communication directed to a first application by a second application, a threat level is computed based at least in part on the information. As a result of the threat level being of a first severity, the second application is migrated to a destination zone that allows for improved communications with the first application. As a result of the threat level being of a second severity, migration of the second application to the destination zone is delayed. As a result of the threat level being of a third severity, a mitigation action is performed.

Distributed denial of service mitigation in a container based framework

Devices, systems, and methods are provided for classifying voice search queries. A system may receive voice data associated with a voice utterance, the system being associated with a network. The system may determine that the voice data is associated with a question, and may determine an absence of an answer to the question. The system may determine a score associated with the question, the score indicative of a risk of disclosure of sensitive information associated with a person. The system may determine that the score fails to satisfy a threshold, and may send the question to a device, wherein the device is remote from the network. The system may receive data associated with the question.

Classifying voice search queries for enhanced privacy

This document describes techniques for placing instances on servers of a service provider environment. In an example embodiment, a system can determine that characteristics of a new virtual machine instance match a co-tenant policy of an existing virtual machine instance on a server and factor that information into a decision to place the virtual machine instance.

Policy-based virtual machine instance cotenancy

Techniques for determining whether data associated with an autonomous/non-autonomous operation of a manned/unmanned vehicle may be trusted. For example, a first set of data may be provided from a source external to a manned/unmanned vehicle. A second set of data may be accessed. This second set may be provided from a source internal or external to the manned/unmanned vehicle and may be associated with the same autonomous/non-autonomous operation. The two sets may be compared to determine whether the first set of data may be trusted or not. If untrusted, a corrective action may be performed.

Detecting of navigation data spoofing based on signal strength variance

Described are techniques for outputting or suppressing output of network content to a user device based on content previously accessed by the user device. Correspondence between network content accessed by the user device and spoiler data determined from media content may indicate that the network content is associated with the media content. Content consumption data associated with the user device may indicate whether the user device has previously accessed the media content. The network content may be suppressed from output if the user device has not previously accessed the media content. The network content may be output to the user device if the media content has been accessed.

System for suppressing output of content based on media access

The present disclosure describes a system and method that identifies locations in source code files that are associated with particular Web requests. The system generates a fingerprint for each Web request based at least in part on the parameters of each HTTP request. By fingerprinting the properties of the source code that generates each HTTP request, Web requests that are generated by executing the fingerprinted code can be traced back to specific source code files, and in some cases an exact line of code. In many examples, a webpage or an action on a webpage can be traced back to a line of source code that is associated with the request. This may allow a developer to find a software defect or security vulnerability by tracing Web requests of a running application and then mapping suspect Web requests back to corresponding lines of code.

Source code mapping through context specific key word indexes and fingerprinting

This disclosure describes systems and processes using multirotor lifter to deploy and/or engage fixed wing aircraft. For example, one or more unmanned multirotor lifters may engage an unmanned fixed wing aircraft, aerially navigate the fixed wing aircraft vertically to a desired altitude, and then release the fixed wing aircraft so that the fixed wing aircraft can initiate a flight plan. In some implementations, multirotor lifter may also be configured to engage fixed wing aircraft while both the multirotor lifters and the fixed wing aircraft are in flight.

Using multirotor lifters to rendezvous with fixed wing aircraft

A non-transitory computer-readable storage device stores instructions that, when executed on a computing system, cause the computing system to receive a request for creating a new software container and determine that characteristics of the new software container match a co-tenant policy of an existing software container on a server. The instructions further cause the computing system to determine that characteristics of the existing software container match a co-tenant policy of the new software container as well as cause the new software container to be created on the server.

Policy-based container cotenancy

Techniques for determining whether data associated with an autonomous operation of an unmanned vehicle may be trusted. For example, a first set of data may be provided from a source external to the unmanned vehicle. A second set of data may be accessed. This second set may be provided from a source internal to the unmanned vehicle and may be associated with the same autonomous operation. The two sets may be compared to determine whether the first set of data may be trusted or not. If untrusted, the autonomous navigation may be directed based on the second set of data and independently of the first set.

Detecting of navigation data spoofing based on sensor data

Techniques for determining whether data associated with an autonomous operation of a first unmanned vehicle may be trusted. For example, the first unmanned vehicle may receive an indication related to the data and originating from a second unmanned vehicle over a network. For instance, the indication may indicate that similar data for a similar autonomous operation of the second unmanned vehicle may be untrusted. Based on a level of trust accorded to the indication, the first unmanned vehicle may determine that the data may be untrusted and the autonomous navigation may be directed independently of the data.

Communication of navigation data spoofing between unmanned vehicles

The subject technology provides embodiments for providing an index of source code repositories isolated to HTTP request parameters. By fingerprinting specific actions involving HTTP requests, such actions can be traced to specific source code files, and in some cases an exact line of code. The end result is that a webpage and any action on that webpage can be traced back to the exact line of source code that is accepting that request. This will allow the ability to find a software defect or security vulnerability while looking at a running application and then map it back directly to the line of code that needs to be modified.

Described are techniques for associating messages with a particular portion of media content. A message received from a first device, associated with a portion of media content stored on the first device, may be provided to a second device and stored in association with a corresponding portion of media content on the second device. Content consumption data associated with the second device may indicate whether the second device has previously accessed the portion of the media content. The message may be suppressed from presentation if the second device has not previously accessed the corresponding portion of the media content. The message may be presented to the second device when the corresponding portion of the media content is accessed.

System for providing messages through media content

Techniques for determining whether data associated with an autonomous operation of an unmanned vehicle may be trusted. For example, the data may be analyzed in light of a capability of the unmanned vehicle. The analysis may indicate an operation of the unmanned vehicle. If the operation is unsupported by the capability, the data may be determined to be untrusted. Accordingly, the autonomous navigation may be directed independently of the untrusted data.

Analyzing navigation data to detect navigation data spoofing

Techniques for determining whether data associated with an autonomous navigation of an unmanned vehicle may be trusted. For example, navigation-related data may be provided from a source external to the unmanned vehicle. Image data associated with the autonomous navigation may be generated. The navigation-related data and the image data may be compared to determine whether the navigation data may be trusted or not. If untrusted, the autonomous navigation may be directed independently of the navigation data.

Growing community of inventors

Mercer Island, WA, United States of America

Scott Gerard Carmack