Search “AWS”

AWS releases a tutorial on distilling a shopping agent with SeqKD on SageMaker AI

On November 3, 2026, the official AWS Machine Learning blog published a tutorial on how to build an efficient shopping agent on Amazon SageMaker AI using sequence-level knowledge distillation (SeqKD). The tutorial uses GLM-5.3 on Amazon Bedrock as the teacher model to generate trajectories, uses the ShoppingBench scorer to retain only trajectories that solve the task, and then performs serverless SFT on the Qwen3.5-4B student model. On 324 held-out test questions, the average absolute success rate of the distilled 4B student rose from 0.297 to 0.541, essentially on par with the teacher model and mid-sized frontier models; when the teacher's reasoning process was removed and only tool calls were learned, the average success rate dropped to 0.415. The tutorial is now publicly available alongside the official blog post.

AWS publishes tutorial on distilling a shopping Agent with SeqKD on SageMaker AI

On November 3, 2026, the official AWS Machine Learning blog published a tutorial on how to build an efficient shopping agent on Amazon SageMaker AI using sequence-level knowledge distillation (SeqKD). The tutorial uses GLM-5.3 on Amazon Bedrock as the teacher model to generate trajectories, uses the ShoppingBench scorer to retain only trajectories that solve the task, and then performs serverless SFT on the Qwen3.5-4B student model. On 324 held-out test questions, the average absolute success rate of the distilled 4B student rose from 0.297 to 0.541, essentially on par with the teacher model and mid-sized frontier models; when the teacher's reasoning process was removed and only tool calls were learned, the average success rate dropped to 0.415. The tutorial is now publicly available alongside the official blog post.

View story:AWS releases a tutorial on distilling a shopping agent with SeqKD on SageMaker AI

HCLTech and AWS launch nano GCC service model

HCLTech launches an AWS-based nano GCC (Global Capability Center) service model for enterprises looking to establish AI operations with small, product-oriented teams; the offering is part of its GCC 360 suite. AWS provides cloud infrastructure, agentic AI services and offerings such as Amazon Bedrock, Kiro and AWS Transform, while HCLTech handles operating model design and implementation, initially rolling out in the Asia-Pacific region. The model targets enterprises that already have GCCs as well as mid-sized companies hoping to enter the Indian market; pricing and customer deployment details were not disclosed.

Andy Jassy reiterates that AWS can become a trillion-dollar revenue business

On October 30, 2026, Amazon CEO Andy Jassy reiterated on the quarterly earnings call that AWS is expected to become a trillion-dollar annual revenue business within the next 5 to 10 years, and he expects growth rates to exceed 40% in both 2027 and 2028. He also said that cash capital expenditures in 2026 will be about $220 billion, most of it directed to AWS, and will increase significantly further in 2027. On the same day, Amazon reported its third-quarter 2026 financial results: net sales rose 12% year over year to $201.94 billion, slightly below expectations; AWS net sales rose 39.3% year over year to $45.98 billion, the highest single-quarter growth rate in nearly five years. Purchases of property and equipment in the quarter reached $77.26 billion, and free cash flow over the past 12 months swung from an inflow of $14.79 billion in the same period a year earlier to an outflow of $42.36 billion; the company maintained its full-year 2026 cash capital expenditure guidance of about $220 billion unchanged. After the results were released, the stock fell by as much as more than 10% in after-hours trading.

Jassy reiterates that AWS can become a trillion-dollar revenue business

On October 30, 2026, Amazon CEO Andy Jassy reiterated on the quarterly earnings call that AWS is expected to become a trillion-dollar annual revenue business within the next 5 to 10 years, and he expects growth rates to exceed 40% in both 2027 and 2028. He also said that cash capital expenditures in 2026 will be about $220 billion, most of it directed to AWS, and will increase significantly further in 2027. On the same day, Amazon reported its third-quarter 2026 financial results: net sales rose 12% year over year to $201.94 billion, slightly below expectations; AWS net sales rose 39.3% year over year to $45.98 billion, the highest single-quarter growth rate in nearly five years. Purchases of property and equipment in the quarter reached $77.26 billion, and free cash flow over the past 12 months swung from an inflow of $14.79 billion in the same period a year earlier to an outflow of $42.36 billion; the company maintained its full-year 2026 cash capital expenditure guidance of about $220 billion unchanged. After the results were released, the stock fell by as much as more than 10% in after-hours trading.

View story:Andy Jassy reiterates that AWS can become a trillion-dollar revenue business

AWS introduces new application and team permission management for QA Studio

On October 29, 2026, the AWS Machine Learning Blog published a post introducing two new capabilities added to QA Studio based on Amazon Nova Act. The first is applications: it groups test cases and test suites into application containers with owners, records CloudWatch metrics by application and environment, and provides a health score from 0 to 100, starting from the pass rate and deducting points for flaky tests, with support for two trend windows: 7 days/hourly and 30 days/daily. The second is team-based user management: it establishes five cumulative roles—Admin, Team Lead, Editor, Executor, and Viewer—where each application belongs to only one team, and permissions are managed by application rather than by individual test. There are currently no follow-up development reports.

AWS introduces enterprise-grade Agentic AI platform blueprint based on AgentCore

On October 29, 2026, AWS released an enterprise-grade Agentic AI platform blueprint based on Amazon Bedrock and AgentCore. The blueprint adopts a federated operating model: a central Platform team provides governed shared services, while each Workstream team retains ownership of its own agents, tools, and data. The platform enforces model admission, regional restrictions, rate and token quotas, and budget controls through an independent inference gateway and MCP tool gateway, and uses an evaluation pipeline to perform gate checks before agents go live. There are currently no subsequent developments.

HCLTech and AWS launch nano GCC service model

HCLTech launches an AWS-based nano GCC (Global Capability Center) service model for enterprises looking to establish AI operations with small, product-oriented teams; the offering is part of its GCC 360 suite. AWS provides cloud infrastructure, agentic AI services and offerings such as Amazon Bedrock, Kiro and AWS Transform, while HCLTech handles operating model design and implementation, initially rolling out in the Asia-Pacific region. The model targets enterprises that already have GCCs as well as mid-sized companies hoping to enter the Indian market; pricing and customer deployment details were not disclosed.

View story:HCLTech and AWS launch nano GCC service model

AWS introduces new application and team permission management in QA Studio

On October 29, 2026, the AWS Machine Learning Blog published a post introducing two new capabilities added to QA Studio based on Amazon Nova Act. The first is applications: it groups test cases and test suites into application containers with owners, records CloudWatch metrics by application and environment, and provides a health score from 0 to 100, starting from the pass rate and deducting points for flaky tests, with support for two trend windows: 7 days/hourly and 30 days/daily. The second is team-based user management: it establishes five cumulative roles—Admin, Team Lead, Editor, Executor, and Viewer—where each application belongs to only one team, and permissions are managed by application rather than by individual test. There are currently no follow-up development reports.

View story:AWS introduces new application and team permission management for QA Studio

AWS introduces enterprise-grade Agentic AI platform blueprint based on AgentCore

On October 29, 2026, AWS released an enterprise-grade Agentic AI platform blueprint based on Amazon Bedrock and AgentCore. The blueprint adopts a federated operating model: a central Platform team provides governed shared services, while each Workstream team retains ownership of its own agents, tools, and data. The platform enforces model admission, regional restrictions, rate and token quotas, and budget controls through an independent inference gateway and MCP tool gateway, and uses an evaluation pipeline to perform gate checks before agents go live. There are currently no subsequent developments.

View story:AWS introduces enterprise-grade Agentic AI platform blueprint based on AgentCore

Dharmendra Raghuvanshi said Chai Point's monthly AWS bill fell 58.2%

Dharmendra Raghuvanshi of Chai Point said the company's monthly AWS bill dropped from $10,877 to $4,550, a 58.2% reduction. According to Analytics India Magazine, about 80% of the savings came from an Agentic Cost Audit performed by a senior engineer using Claude Code: the audit examined production and QA cloud accounts in read-only mode, identifying idle and orphaned resources, over-provisioning and misconfigurations, and ranked them by savings amount, implementation cost and risk. The remaining roughly 19% came from switching AWS reseller to Isaga for a 5% discount, and from Savings Plans reducing committed spending by about 15%. The report said the platform still supports more than 8,000 connected brewing devices across over 170 stores.

AWS launches SOCI-indexed images for SageMaker AI jobs

On October 28, 2026, AWS launched SOCI-indexed images for Amazon SageMaker AI training and processing jobs, enabling containers to start within seconds through lazy loading without requiring changes to job definitions. Official benchmarks published by AWS Machine Learning show that download times for images ranging from 1.8–35.4 GB dropped from 48 seconds to 7 minutes 17 seconds down to 3–11 seconds, a speedup of 8x–46x, with GPU idle time reduced by up to 90%. If the index is unreadable or lazy loading fails, jobs automatically fall back to standard downloads. There are currently no further developments.

AWS launches SOCI-indexed images for SageMaker AI jobs

On October 28, 2026, AWS launched SOCI-indexed images for Amazon SageMaker AI training and processing jobs, enabling containers to start within seconds through lazy loading without requiring changes to job definitions. Official benchmarks published by AWS Machine Learning show that download times for images ranging from 1.8–35.4 GB dropped from 48 seconds to 7 minutes 17 seconds down to 3–11 seconds, a speedup of 8x–46x, with GPU idle time reduced by up to 90%. If the index is unreadable or lazy loading fails, jobs automatically fall back to standard downloads. There are currently no further developments.

View story:AWS launches SOCI-indexed images for SageMaker AI jobs

Chai Point says auditing with Claude Code cut its monthly AWS bill by 58.2%

Dharmendra Raghuvanshi of Chai Point said the company's monthly AWS bill dropped from $10,877 to $4,550, a 58.2% reduction. According to Analytics India Magazine, about 80% of the savings came from an Agentic Cost Audit performed by a senior engineer using Claude Code: the audit examined production and QA cloud accounts in read-only mode, identifying idle and orphaned resources, over-provisioning and misconfigurations, and ranked them by savings amount, implementation cost and risk. The remaining roughly 19% came from switching AWS reseller to Isaga for a 5% discount, and from Savings Plans reducing committed spending by about 15%. The report said the platform still supports more than 8,000 connected brewing devices across over 170 stores.

View story:Dharmendra Raghuvanshi said Chai Point's monthly AWS bill fell 58.2%