Amazon Technologies, Inc.

Jon Arron McClintock

Darren Ernest Canavor

Jesper Mikael Johansson

George Nikolaos Stathakopoulos

Tushaar Sethi

Brandon William Porter

Andrew Jay Roths

Daniel Wade Hitchcock

Maarten Van Horenbeeck

Gregory Branchek Roth

Bharath Kumar Bhimanaik

Dominique Imjya Brezinski

Nima Sharifi Mehr

Harsha Ramalingam

Varadarajan Gopalakrishnan

Karl Andre McCabe

David Erdmann

Matthew Ryan Jezorek

Alun Jones

Katharine Nicole Harrison

Gregory Alan Rubin

Darin Keith McAdams

Narasimha Rao Lakkakula

Thibault Candebat

Christopher Michael Anderson

Arnaud Marie Froment

Chelsea Celest Krueger

Adam Cecchetti

Phivos Costas Aristides

Eric J Martin

Donghui Zhuo

Scott Donald Gregory

Vijay Rangarajan

Cory Adam Johnson

Jonathan Kozolchyk

Yogesh Vilas Golwalkar

Muhammad Wasiq

James Connelly Petts

Max Harwell Funderburk

Robert Ellis Lee

Jason Christopher Rudmann

David Scott Cleckley, Jr

Annabelle Richard Backman

Dylan Harris Rush

Zachary Damen Wolfe

Brian Dang

Eric Jason Brandwine

Daniel Buchmueller

Fabian Hensel

Scott Gerard Carmack

Melissa Elaine Davis

Anton Vladilenovich Goldberg

William Frederick Kruse

Sachin Purushottam Joglekar

John Clark Coonley Duksta

David James Roy Erdmann

John Elias Darrow

James Domit Mackraz

Brian Young Lee

Kadirvel Chockalingam Vanniarajan

Ketan Ramesh Deshpande

Aram Grigoryan

Dean H Saxe

David John Burke

Patrick Nicholas Denton

Conor Power

Sunu Aby Mathew

Shailendra Batra

Assim Deodia

Nathan David Scherer

Danial Muhammad Ranjha

Mayur Sharma

Matthew James Parker

Matthew Paul Wenger

Preyas Joshi

Techniques described and suggested herein include various systems and methods for determining risk levels associated with transiting data, and routing portions of the data in accordance with the determined risk levels. For example, a risk analyzer may apply risk classifiers to transiting data to determine overall risk levels of some or all of the transiting data. A traffic router may route transiting data according to determined risk profiles for the data. A sandbox may be implemented to compare, for a given input, expected and observed outputs for a subset of transiting data, so as to determine risk profiles associated with at least the subset.

Techniques for data routing and management using risk classification and data sampling

Disclosed are various embodiments for tracking developer behavior with respect to software analysis tools. In one embodiment, a security analysis is performed upon a first revision of a program, where the security analysis is based at least in part on a plurality of rules. A first security issue found in the security analysis upon the first revision of the program is identified. The security analysis is performed upon a second revision of the program. A second security issue found in the security analysis upon the second revision of the program is identified. The rules are updated based at least in part on whether the first security issue is corrected in the second revision as determined based at least in part on a comparison of the first security issue to the second security issue.

Tracking developer behavior with respect to software analysis tools

Pattern based detection of data usage is facilitated using data injection. Data values are injected in one or more storage locations accessible to a plurality of services or included in service requests. Service interactions among the services are compared to a set of patterns. The set of patterns are configured to match the data values. By comparing the service interactions to the patterns, one or more of the service interactions are determined to include individual ones of the data values. Data are generated indicating a presence of the data values in the services.

Pattern-based detection using data injection

Techniques are described for determining a priority order for generating, serving, or rendering components of content such as a web page. Behavioral data may be collected from user devices, the behavioral data describing user interactions with components of the content during previous presentations of the content on the user devices. Based on the behavioral data, a score may be determined for one or more of the components, the score based on component dwell times, component presentation frequency, or other information. A priority order for the components may be based on the determined scores, and the components may be generated, served, or rendered in the priority order.

Behavior based optimization for content presentation

Methods and systems are provided for managing access to a client account related (CAR) resource. When a privilege-constrained (PC) application requests access to an individual client account, a single use authorization (SUA) code is created that is associated with the individual client account. The SUA code is routed to, and returned from, the privilege-constrained (PC) application to authenticate the PC application. The PC application, once authenticated, receives a permitted action token that identifies a limited set of privileges that the PC application is authorized to perform in connection with the CAR resource. The PC application provides the permitted action token to an access service. The access service limits access, by the PC application, to the CAR resource based on the permitted action token.

Authentication and authorization of a privilege-constrained application

Two unmanned vehicles come within communication range of one another. The unmanned vehicles exchange logs of messages each has received. Each of the unmanned vehicles analyzes the messages that it received from the other unmanned vehicle to determine whether any of the received messages warrants changing a set of tasks it was planning to perform. When a message indicates that a task should be changed, the task is updated accordingly.

Unmanned vehicle message exchange

Disclosed are various embodiments for malware detection by way of proxy servers. In one embodiment, a proxied request for a network resource from a network site is received from a client device by a proxy server application. The proxied request is analyzed to determine whether the proxied request includes protected information transmitted in an unsecured manner. It is then determined whether the network resource comprises malware based at least in part on an execution of the network resource or whether the proxied request includes the protected information transmitted in the unsecured manner. The proxy server application refrains from sending data generated by the network resource to the client device in response to the proxied request when the network resource is determined to comprise the malware.

Proxy server-based malware detection

Disclosed are various embodiments for validating that relying parties of a federated identity provider have correctly implemented sign-out functionality. In one approach, a network page is received from a network site that is operated by a relying party of a federated identity provider. It is then determined whether the network page includes code that properly implements a sign-out from the federated identity provider. An action is initiated in response to determining that the network page does not include code that properly implements the sign-out from the federated identity provider.

Validating sign-out implementation for identity federation

A client establishes a network session with a server. The network session is used to establish an encrypted communications session. The client establishes another network session with another server, such as after terminating the first network session. The client resumes the encrypted communications session over the network session with the other server. The other server is configured to receive encrypted communications from the client and forward them to the appropriate server.

Techniques for secure session reestablishment

A method and apparatus for risk scoring in a graph are disclosed. In the method and apparatus, a graph includes a first node that is connected with a node of a plurality of nodes using a communication link of a plurality of communication links. A plurality of link risk measures are then determined, whereby a link risk measure of the plurality of link risk measures pertains to the communication link of the plurality of communication links. Furthermore, a risk measure associated with the first node is determined based at least in part on the plurality of link risk measures. The risk measure is monitored to determine if one or more conditions placed on the risk measure are met and one or more actions are taken as a result of the one or more conditions being met.

Risk scoring in a connected graph

Disclosed are various embodiments for transitioning user accounts from an assisted status to a self-managed status. In one embodiment, a service receives an authentication request from a client device, where the authentication request specifies a unique user identifier corresponding to a communication channel. The service then determines that the unique user identifier corresponds to an assisted user account but the client device is not logged in under an assisting user account. The service generates an ephemeral security credential and sends the ephemeral security credential to the communication channel. The service then creates a self-managed user account having the unique user identifier in response to receiving the ephemeral security credential from the client device.

Migrating user accounts from assisted to self-managed

The current document describes systems and methods that provide access controls in a system of interconnected services such as an online service platform. In various implementations, the system maintains contextual information associated with tokenized data. In additional implementations, data brokers authorize access to detokenized data by comparing the context of the data to the context of the service requesting the data. In yet additional implementations, the system maintains contextual information associated with requests that are processed within the system. When a request is made to a particular service, the particular service can use the identity of the requester, the context of the request, and the context of the data to determine whether the request is authorized. In some implementations, the integrity of contextual information is protected using a digital signature.

Context-based data access control

A system and method for activating security mechanisms based at least in part on accelerometer-based dead reckoning wherein accelerometer data, reflecting acceleration in a local coordinate system of a device, is obtained from an accelerometer of a device. Movement of the device is determined based at least in part on the accelerometer data, and, based at least in part on whether the movement of the device exceeds a threshold value, a determination is made whether to change a current security state of the device. If it is determined to change the current security state of the device, the current security state of the device is changed to a new security state.

Activation of security mechanisms through accelerometer-based dead reckoning

A method and apparatus for device authentication are provided. In the method and apparatus, authentication data for a first device is received. The first device is then authenticated based at least in part on demonstrated access to authentication data prior to broadcast of the authentication data. One or more actions may be taken in response to the authentication of the first device based at least in part on the demonstrated access to the authentication data.

Broadcast-based trust establishment

Growing community of inventors

Seattle, WA, United States of America

Jon Arron McClintock