AI data platform iMerit believes the next step toward integrating AI tools at the enterprise level is not more data, but better data. The startup has quietly built itself into a trusted data annotation partner for companies working in computer vision, medical imaging, autonomous mobility, and other AI applications that require high-accuracy, human-in-the-loop labeling. Now, iMerit is bringing its Scholars program out of beta. The goal of the program is to build a growing workforce of experts to fine-tune generative AI models for enterprise applications and, increasingly, foundational models. iMerit doesn’t claim to replace Scale AI’s core offering of high-throughput, developer-focused “blitz data.” Instead, it’s betting that now is the right moment to double down on expert-led, high-quality data, the kind that requires deep human judgment and domain-specific oversight. iMerit’s experts are tasked with finetuning, or “tormenting,” enterprise and foundational AI models using the startup’s proprietary platform Ango Hub. Ango allows iMerit’s “Scholars” to interact with the customer’s model to generate and evaluate problems for the model to solve. For iMerritt, attracting and retaining cognitive experts is key to success because the experts aren’t just doing a few tasks and disappearing; they’re working on projects for multiple years. The goal is to grow across other enterprise applications, including finance and medicine.
Penske Logistics taps Snowflake AI to develop AI-based program that flags drivers at risk of quitting based on work patterns, route history and behavioral signals
Penske Logistics has leveraged Snowflake Inc.’s evolving artificial intelligence capabilities through a strategic partnership that’s reshaping the supply chain landscape. “We have onboard telematics devices inside our fleet that are generating millions of data points, including things like hard braking, following too closely, fuel consumption and so on,” said Vishwa Ram, vice president of data science and analytics at Penske Logistics. “Getting all of that data in one place and adding it up with other sets of data that we have that are contextual is a huge challenge for us.” “We’re accustomed now to disruption being normal, and as a result, organizations see just how important it is to invest in that visibility element so they can see the disruption as it’s coming, or at least be able to react in real time when it does happen,” he said. For Penske, this means leveraging predictive analytics to foresee supplier delays and reroute resources before bottlenecks occur. Additionally, the company applies AI in workforce retention. With drivers making up over half of its workforce, driver satisfaction is key. Penske developed an AI-based program that flags drivers at risk of quitting based on work patterns, route history and behavioral signals. Armed with these insights, frontline managers proactively engage with drivers, often adjusting schedules or simply checking in, according to Ram.
Arctera.io’s backup and storage solution combines data management, cyber resiliency and data compliance and is designed to monitor customers’ data environments for potential security breaches by tracking deterministic AI, that changes over time
Arctera.io is going full steam ahead on data and artificial intelligence management. Arctera offers three backup and storage options that originated from Veritas Technologies LLC, covering data management, cyber resiliency and data compliance. All three areas are under a microscope in the era of AI adoption. Arctera focuses on data management that meets regulatory requirements and remains secure against ongoing threats from cyberattackers. AI presents a particular challenge to data resilience and recovery because it’s constantly changing, according to Matt Waxman, chief product officer at Arctera.io. “We have built IT around the notion that software is deterministic,” he explained. “The notion that [this] is static, that the software that you’re acquiring is going to be the same software at least for quite a long period of time. That’s not the case with AI. So what you bring in terms of a model is self-learning, and it’s going to adjust over time.” For this reason, it’s crucial to have multiple ways to back up your data and multiple ways to keep track of it. While Waxman advises against “AI whitewashing,” or applying AI to problems indiscriminately, Arctera has successfully implemented use cases for AI related to data compliance and monitoring. Arctera attempts to kill two birds with one stone by employing its data compliance software to monitor customers’ data environments for potential security breaches. The next step is governing the growing host of AI agents, according to Waxman.
Indico unveils embedded data enrichment agents to supercharge insurance decisioning by transforming unstructured submissions, claims, and policy documents into structured, decision-ready data
Indico Data has expanded its Data Enrichment Agents, enhancing document workflows with deeper, native access to proprietary and third-party datasets. The enrichment capabilities combine Indico’s growing library of proprietary data catalogs with seamless integration to proprietary data and trusted third-party providers. The available data spans commercial, personal, and property domains, and includes enriched details such as business credit and risk scores, crime statistics, driver safety and motor vehicle violations, VIN and registration data, proximity-based risk, co-tenancy exposure, property characteristics, permit activity, and more. The Data Enrichment Agents are now generally available to all Indico platform customers and can be activated across workflows including submission ingestion, underwriting clearance, claims FNOL, and policy servicing. By transforming unstructured submissions, claims, and policy documents into structured, decision-ready data, Indico enables insurers to act faster on high-value opportunities, streamline triage and intake, and improve the consistency and transparency of underwriting and claims decisions. Key benefits of Indico’s Data Enrichment Agents include: Embedded data access; Auto-fill missing data; Flexible provider ecosystem; and Proprietary data at a lower cost.
HighByte’s Industrial MCP Server enables AI agents to securely access all connected industrial systems and make real time or historical data requests on them by exposing data pipelines as “tools” and including descriptions and parameters
HighByte has released HighByte Intelligence Hub version 4.2 with an embedded Industrial Model Context Protocol (MCP) Server that powers Agentic AI and new LLM-assisted data contextualization via native connections to Amazon Bedrock, Azure OpenAI, Google Gemini, OpenAI, and local LLMs. HighByte Intelligence Hub provides the first Industrial MCP Server to expose data pipelines as “tools” to AI agents, including descriptions and parameters. With the Intelligence Hub, AI agents can securely access all connected industrial systems and make real time or historical data requests on them. John Harrington, Chief Product Officer at HighByte said “The Intelligence Hub is an Industrial DataOps solution that contextualizes and standardizes industrial data from diverse sources for diverse targets. Agentic AI clients on the factory floor are a natural extension of this approach. We’re enabling DataOps to feed AI, and AI to assist and scale DataOps.” The latest release also introduces Git integration and OpenTelemetry (OTel) support to scale and manage deployments using DevOps tooling for version control and observability. Users will also have access to new Databricks and TimescaleDB connectors and enhanced connectivity with Apache Kafka and Amazon S3 for cloud-to-edge use cases. Furthermore, the Oracle Database connection has been enhanced to support Change Data Capture (CDC), the Snowflake SQL connection now supports write operations, and the AVEVA PI System connection supports enhanced PI point metadata reads. These capabilities optimize bi-directional connectivity for the many disparate data services found in the cloud, data center, and factory floor.
Bright Data’s AI browser targeted at AI agents runs in the cloud, supports natural language prompts, bypasses CAPTCHAs, scripts, and bot defenses and mimics real user behavior to access and interact with the web at scale
Bright Data, the world’s #1 web data infrastructure company for AI & BI, has launched a powerful set of AI-powered web search and discovery tools designed to give LLMs and autonomous agents frictionless access to the open web: Deep Lookup (Beta): Deep Lookup (Beta) is a natural language research engine that answers complex, multi-layered questions in real-time, with structured insight. Deep Lookup (Beta) allows users to query across petabytes of unstructured and structured web data simultaneously, surfacing high-confidence answers to complex, multi-layered questions, without code. Unlike general-purpose LLMs that hallucinate or struggle with context, Deep Lookup (Beta) delivers verified, web-sourced insights, with links to cited sources, with structured outputs you can immediately act on—across thousands of verticals. Browser.ai: The industry’s first unblockable, AI-native browser. Designed specifically for autonomous agents, Browser.ai mimics real user behavior to access and interact with the web at scale. It runs in the cloud, supports natural language prompts, and bypasses CAPTCHAs, scripts, and bot defenses, making it ideal for scaling agent-based tasks like scraping, monitoring, and dynamic research. MCP Servers: A low-latency control layer that lets agents search, crawl, and extract live data in real-time. Built to power agentic workflows, MCP is designed for developers building Retrieval-Augmented Generation (RAG) pipelines, autonomous tools, and multi-agent systems that need to act in context, not just passively read.
Structify’s AI platform combines visual language model with human oversight to simplify data preparation by letting users create custom datasets by specifying the data schema, selecting sources, and deploying AI agents to extract that data through navigating the web
Startup Structify is taking aim at one of the most notorious pain points in the world of artificial intelligence and data analytics: the painstaking process of data preparation. The company’s platform uses a proprietary visual language model called DoRa to automate the gathering, cleaning, and structuring of data — a process that typically consumes up to 80% of data scientists’ time. At its core, Structify allows users to create custom datasets by specifying the data schema, selecting sources, and deploying AI agents to extract that data. The platform can handle everything from SEC filings and LinkedIn profiles to news articles and specialized industry documents. What sets Structify apart, is their in-house model DoRa, which navigates the web like a human would. This approach allows Structify to support a free tier, which will help democratize access to structured data. Structify’s vision is to “commoditize data” — making it something that can be easily recreated if lost. Finance teams use it to extract information from pitch decks, construction companies turn complex geotechnical documents into readable tables, and sales teams gather real-time organizational charts for their accounts. A key differentiator for Structify is its “quadruple verification” process, which combines AI with human oversight. This approach addresses a critical concern in AI development: ensuring accuracy. What differentiates Structify, according to CEO Alex Reichenbach, is its combination of speed and accuracy. Reichenbach claimed they had sped up their agent “10x while cutting cost ~16x” through model optimization and infrastructure improvements.
Electron AI, the agentic assistant for data teams and analysts generates precise, context-aware mapping logic across source systems, semantic models, and destination schemas
Reactor Data announced the production launch and immediate availability of Electron AI – the embedded conversational AI assistant designed to help data teams and analysts create powerful data mappings, transformations and pipelines. Electron acts as an intelligent co-pilot, enabling data analysts and teams to generate precise, context-aware mapping logic across source systems, semantic models, and destination schemas – all through simple conversational interactions. Electron acts as a natural-language assistant familiar with all aspects of a company’s data pipelines including sources, source schemas, multi-step transformations including complex data combinations, output configurations and destination tables. Whether a business is normalizing product titles, mapping transactional IDs, or aligning common fields across disparate sources, Electron helps brands go from request to result faster, with less friction and fewer mistakes. Key Capabilities of Reactor Data’s Electron AI: Conversational and Multilanguage Coding: Ask Electron to write complex data transformations, and it returns both Python code and simple natural language expressions. Pipeline and Context-Aware: Electron is tightly integrated with Reactor’s modular pipeline tools for source, semantic, and destination processing. Electron understands source and destination schemas and rules to offer precise, pre-validated mappings. Iterative Authoring: Electron translates natural language into mapping expressions with null handling, coalescing, formatting, and refinement based on feedback.
FICO’s new Marketplace to connect enterprises to providers of data, AI/ML models, optimization tools and decision rulesets and cut the time required to access, validate and integrate new data sources
FICO has introduced a digital hub designed to connect organizations with data and analytics providers. This innovative new Marketplace offers easy access to data, artificial intelligence (AI) models, optimization tools, decision rulesets, and machine learning models, which deliver enterprise business outcomes from AI. With FICO Marketplace, FICO® Platform users can fast-track their journey to becoming an intelligent enterprise, because they will be able to: Unlock Value from Data Faster: by experimenting with new data sources and decision assets to determine predictive power and business value. Users can expect to cut the time required to access, validate and integrate new data sources by half. Leverage Decision Agents Across Multiple Use Cases, Improving Collaboration: with its open API architecture, it allows for any decision asset, data service, analytics model, software agent or third-party solution to address a wide range of use cases including customer management, fraud, originations, and marketing. The reusability of decision agents across multiple departments breaks down silos and improves collaboration. Drive Better Customer Experiences: by enabling a holistic view of each individual customer, as well as building innovative new intelligent solutions and analytic capabilities that come from industry collaboration. “FICO Marketplace will facilitate the type of collaboration across the industry that drives the next generation of intelligent solutions,” said Nikhil Behl, president, Software, FICO.
TensorStax’s data engineering AI agents can design and deploy data pipelines through structured and predictable orchestration using a deterministic control layer that sits between the LLM and the data stack
Startup TensorStax is building AI agents that can perform tasks on behalf of users with minimal intervention to the challenge of data engineering. The startup gets around this by creating a purpose-built abstraction layer to ensure its AI agents can design, build and deploy data pipelines with a high degree of reliability. Its proprietary LLM Compiler acts as a deterministic control layer that sits between the LLM and the data stack to facilitate structured and predictable orchestration across complex data systems. Among other things, it does the job of validating syntax, normalizing tool interfaces and resolving dependencies ahead of time. This helps to boost the success rates of its AI agents from 40% to 50% to as high as 90% in a variety of data engineering tasks, citing internal testing. The result is far fewer broken data pipelines, giving teams the confidence to offload various complicated engineering tasks to AI agents. TensorStax says its AI agents can help to mitigate the operational complexities involved in data engineering, freeing up engineers to focus on more complex and creative tasks, such as modeling business logic, designing scalable architectures and enhancing data quality. By integrating directly within each customer’s existing data stack, TensorStax makes it possible to introduce AI agent data engineers into the mix without disrupting workflows or rebuilding their data infrastructure. These agents are designed to work with dozens of common data engineering tools. The best thing is that TensorStax AI agents respond to simple commands. Constellation Research Inc. analyst Michael Ni said TensorStax appears to be architecturally different to others, with its LLM compiler, its integration with existing tools and its no-customer-data-touch approach.