Imagine a batch job that has run every night for decades without failing, and the only person who understands its code is about to retire. We have seen this situation again and again while working with banking, insurance, and airline customers. Their Assembler programs still run critical business operations, but fewer people can read them each year. Rewriting them by hand takes years, risks ...
John Urschel PhD ’21, the Class of 1956 Career Development Assistant Professor in the MIT Department of Mathematics, is no stranger to building muscle: Before earning his PhD at MIT, Urschel strengthened his offensive muscles as a lineman for the Baltimore Ravens. He remains committed to helping his students exercise, but now his focus is on critical thinking skills. In the classroom, U...
With Denied Topics in Amazon Bedrock Guardrails, you block a content category by describing it in plain language, and the guardrail applies that description to every request by meaning rather than by keyword. That same mechanism can also match more than you intend. A topic named Finance, meant to keep an assistant out of investment advice, also blocks “When does my direct deposit clear? ”. I...
In its MLPerf benchmarks debut, with results released today, the NVIDIA Vera Rubin platform delivers leading performance across training, post-training and agentic inference. NVIDIA swept every MLPerf Training v6. 1 benchmark and, in its first MLPerf Endpoints submission, delivered more than 10x generational gains in throughput per watt for agentic AI — measuring how much agentic work a syst...
The goal of a machine learning binary classification system is to predict the value of a variable that can have two discrete values. For example, you might want to predict the sex (male = 0, female = 1) of a person based on age, State of residence, income and political leaning (conservative, moderate, liberal). There are approximately a dozen binary classification techniques. Techniques incl...
Financial institutions face a critical challenge: processing massive volumes of data with speed, accuracy, and compliance while operating under some of the world’s strictest regulatory frameworks. Duco is the agentic Operations platform for financial services, serving tier-one banks and other financial institutions. It processes 20 billion transactions monthly for 200+ clients, including 7 o...
Enterprise teams are building multi-agent systems, and when early implementations fall short, the cause is rarely the model. Multi-agent coordination is a distributed systems problem, and many teams taking it on haven’t yet built distributed systems instincts. The numbers are consistent across sources: ~88% of AI pilots never reach production. Stanford HAI demonstrated that two models can pe...
Novo Nordisk (Novo) is one of the world’s leading global pharmaceutical companies, manufacturing medicines for chronic diseases such as diabetes and obesity, and delivering them to patients across many countries. Behind every product lies a manufacturing network that adheres to strict quality requirements. Every quality-relevant process step must be described, controlled, and validated throu...
This post was co-written with Ashwin Srinivasan, Nishan Das and Karthik Baile from MoEngage. A marketing or product team member planning a weekend flash sale often opens three tools before writing a single line of copy. First they open the analytics dashboard to see what worked last time, then the content library to find the right tone, and then the campaign builder to build the campaign its...
When deploying deep learning models in high-stakes scenarios, researchers would like to understand how they work. Today, this requires unraveling an enormously complex network of internal processing nodes to understand what the model knows and how it will likely behave in practice. Jacob Andreas would rather have the models explain themselves. The newly tenured associate professor deve...
Logistic regression is a technique for binary classification. For example, you might want to predict the sex of a person (0 = male, 1 = female) based on age, State, income, and political leaning. I enjoy writing code, so one rainy evening (I live in the Pacific Northwest so this wasn’t unusual), while I was waiting for the rain to stop so I could walk my dogs, I decided to implement lo...
Amazon Bedrock provides built-in, zero-instrumentation cost attribution that traces every inference call back to the IAM principal that made it. The identity flows along with organizational tags, into AWS Cost and Usage Reports (CUR 2. 0) and AWS Cost Explorer while keeping data inside the AWS security boundary. This post shows you how to use these capabilities to answer three questions that...
Speech-to-text inference is computationally intensive. Each audio file triggers a pipeline of deep learning models (acoustic models, language models, and post-processing stages) running on NVIDIA GPUs to meet latency requirements. At scale, this means thousands of GPU instances running continuously. AssemblyAI needed a way to run GPU-intensive inference workloads cost-effectively at scale, w...
From pediatric research to smart hospitals, AI is helping care teams spend less time on routine tasks and more time with patients. Health systems and startups are putting it to work across healthcare, from matching patients to clinical trials to supporting radiologists, hospital teams and patients at home. Clinicians stay at the center, reviewing AI outputs and guiding decisions, while NVIDI...
Agents today review code, process claims, and run unattended for hours. Many also sit behind interactive experiences where a person waits for the first response. Amazon Bedrock AgentCore runtime gives these agents a fully managed, serverless environment: each session runs in its own isolated microVM, the platform scales to zero when there’s no work, and you only pay for the resources your se...