Ex-big tech cyber leaders launch Dawnguard from stealth with $3M to rewrite DNA of cybersecurity

Posted in Commentary with tags on July 31, 2025 by itnerd

 Dawnguard, a cybersecurity startup on a mission to make the digital world safer through intelligent, design-first security, has emerged from stealth with $3 million in pre-seed funding. The round was led by 9900 Capital and a group of angel investors, from scale-up founders to experienced CIOs and CISOs. The funds will be used to expand Dawnguard’s engineering team, deepen enterprise integrations, and bring its platform to broader production use.

Dawnguard is introducing a new cybersecurity category. Rather than bolting on security in production, Dawnguard embeds it at the core of system architecture — ensuring secure, compliant, and scalable designs from the earliest phases of development.

Dawnguard’s holistic approach sets it apart. Rather than just scanning deployments or automating reviews, it provides a shared canvas for engineering and security teams to collaborate on secure, compliant architecture that also balances cost, resilience, and sustainability.

Dawnguard was born out of a broken model – where security was reactive, slow, and dangerously disconnected from the pace of modern development. The founding team, led by CEO Mahdi Abdulrazak and CTO Kim van Lavieren, is composed of industry veterans from military and big tech companies like IBM, Microsoft and Amazon, with decades of experience running large-scale security programs and with unique experience at the intersection of security, AI, and cloud. 

Dawnguard is set to flip shift-left and security-by-design on its head. Instead of treating security as an afterthought, Dawnguard embeds it directly into a system’s architecture, from day zero to day 10,000. The company is building various AI/ML-driven engines that integrate across the entire IT landscape to spot issues in the design phase, adapt to evolving environments, and make security native.

The platform is designed for security architects, DevOps engineers, and cloud teams. At its core, Dawnguard is a security architecture automation platform purpose-built for cloud-native environments. It helps teams validate cloud infrastructure designs before deployment, automatically generate production-ready Infrastructure as Code (IaC) from validated designs, and continuously enforce security posture after deployment to eliminate drift.

“Hundreds of security tools overwhelm CISOs with promises of better detection, yet few tackle the root issue: design flaws in code that AI-driven threats exploit. As attacks grow smarter, defenses must shift left—embedding resilience at the codebase. We are excited to back Dawnguard, who build protection by design, not patch by necessity,” said Chris Corbishley, Managing Partner 9900 Capital.

Looking ahead, Dawnguard aims to reshape how the industry embeds security in the AI era. The company plans to expand its platform to support more dynamic environments, close the security gap between “vibe coding” and the infrastructure where GenAI coded applications run, and deliver a new operating model for building trust at scale.

Leave a comment »

Government Ransomware Roundup: Attacks on the Incline for H1 2025?

Posted in Commentary with tags on July 31, 2025 by itnerd

Comparitech researchers have released a study looking at all the government ransomware attacks of the first half of 2025. Government organizations are a dominant focus for hackers, due to the sensitive information the industry holds.

According to the findings, there was a 65% increase in attacks compared to the first half of 2024.

This research looks at ransomware attacks on government organizations by country, finding the U.S. to have been the most targeted. Additionally, the study outlines the biggest ransomware demands on governments, worldwide, as well as which gangs are most prolific in this industry. 

Key findings include: 

  • 208 attacks in total – 124 in Q1 and 84 in Q2
  • 104 confirmed attacks – 54 in Q1 and 50 in Q2
  • 104 unconfirmed attacks – 70 in Q1 and 34 in Q2
  • 366,006 records are known to have been breached in the confirmed attacks
  • 78.5 TB of data allegedly stolen (67.2 TB in the confirmed attacks)
  • Average theft of 1.3 TB of data per attack
  • Average ransom demand of $1.65 million
  • The most prolific ransomware strains with the highest number of claims against government entities were Babuk (26), Qilin (17), INC (16), Funksec (12), and RansomHub (12)
  • Qilin had the most confirmed attacks (13), followed by INC and RansomHub (8 each)

You can read the research here: https://www.comparitech.com/news/government-ransomware-roundup-h1-2025-stats-on-attacks-ransoms-and-data-breaches/

Leave a comment »

New BEC in the Financial Services Sector Threat Report Finds Nearly 4000 Malicious Domains in Q2 2025 

Posted in Commentary with tags on July 31, 2025 by itnerd

BforeAI has released its new threat report analyzing BEC in the financial services sector, finding 3756 suspicious and newly registered domains in April, May, and June of this year.

In Q2 2025, BforeAI observed 

  • Top 3 registrars: GoDaddy.com, LLC, Dynadot Inc., Tucows Domains Inc.
  • Top 3 registering countries: United States, China, United Kingdom
  • Top 5 TLDs (Top Level Domains): .com (1992), .info (260), .xyz (203), .online (105), .icu (104)
  • Finance-based TLDs: .finance, .financial, .money, .loan, .cash, .fund, .credit, .cards, .accountant, .bank, .investments, .capital, .exchange, .market, .insurance

Domain registration trends throughout the quarter showed a notable spike in activity, especially targeting financial brands. 

April saw a high volume of registrations, followed by a slight dip in May, and then a sharp rise in June, especially towards the end of the month. 

June also recorded the highest number of domain registrations overall. Between June 22 and June 30, 2025, there were at least 22 domains registered daily, peaking at 81 registrations on June 27 alone.

Beyond this, a consistent count of 10 or more newly registered domains was observed daily, with fluctuations continuing through the end of the quarter.

This sudden surge could indicate a sign of preparation for upcoming seasonal retail sales or early travel-related promotions, during which many financial institutions roll out offers and rewards, making this period a prime target for cybercriminals looking to spoof legitimate offers.

You can read the report here: https://bfore.ai/report/bec-in-the-financial-services-sector/

Leave a comment »

Deepgram Expands Internationally, Launches Managed Single-Tenant Deployment Option

Posted in Commentary with tags on July 30, 2025 by itnerd

Voice AI is rapidly becoming foundational infrastructure across industries, powering real-time agents, compliance-sensitive workflows, and multilingual applications at scale. As global adoption accelerates, so does the demand for flexible deployment models, regional hosting, and production-grade reliability.

To meet that demand, Deepgram is announcing two major infrastructure expansions:

  • The general availability of Deepgram Dedicated, a fully managed, single-tenant runtime
  • The early access launch of our EU-hosted API endpoint, enabling in-region inference for European workloads

These launches reflect a broader shift in how voice AI is being deployed, and they come at a time of growing industry validation. This month, Deepgram Nova-3 was named a 2025 Voice AI Technology Excellence Award winner by TMC’s CUSTOMER magazine, recognizing their leadership in accuracy, real-time multilingual transcription, and self-serve customization.

Together, these milestones reinforce Deepgram’s commitment to providing voice AI infrastructure that supports enterprise-scale performance, compliance, and geographic flexibility.

What It Means to Go Global with Voice AI

Going global starts with supporting the world’s languages. Deepgram already supports over 36 languages for customers worldwide and will continue expanding language coverage throughout 2025. 

But language support is only the beginning.

For engineering teams building production-grade systems, global voice AI also requires solving for infrastructure and compliance demands as workloads expand across regions. As enterprises scale voice workloads globally, we continue to hear two common friction points: the growing complexity of managing infrastructure across regions and tightening data policies, particularly in the EU, that require stricter control over where and how voice data is processed.

These demands include:

  • Ultra low-latency inference paths. Real-time applications require models to run as close to the end user as possible to minimize round-trip time and meet interaction thresholds.
  • Data residency and legal jurisdiction. Voice data often must be processed and stored within specific geographic boundaries to meet regulatory requirements such as GDPR.
  • Single-tenant isolation for sensitive workloads. Some environments require dedicated infrastructure to enforce data segregation, meet compliance standards, or satisfy internal security policies.
  • Scalable operations without added DevOps burden. Expanding voice workloads across regions should not require a proportional increase in infrastructure engineering.

Deepgram’s platform was designed with these requirements in mind, providing the foundation needed to operationalize voice AI reliably and securely across global environments.

Introducing Deepgram Dedicated: A Managed, Single-Tenant Runtime

Enterprises adopting voice AI at scale often face a difficult tradeoff: maintain control over infrastructure and data by self-hosting, or prioritize ease of use through shared, multi-tenant cloud APIs. Self-hosting offers isolation and regional control, but introduces significant ongoing operational complexity. Managed service providers can help bridge the gap, but they often lack product-level expertise and introduce dependency overhead that slows down feature adoption.

Now generally available, Deepgram Dedicated closes this gap. It is a fully managed, single-tenant deployment of Deepgram’s voice AI platform that offers the control and flexibility of self-hosted infrastructure without the burden of operating it. Over the past six months, it has been selectively deployed with a select group of enterprise customers in early production deployments across a range of use cases, from real-time contact center platforms to globally distributed voice agents.

Teams gain regional isolation, performance control, and compliance alignment while offloading infrastructure management to Deepgram. Deepgram Dedicated currently runs on AWS, with support for additional cloud providers on the roadmap.

Key Highlights:

  • Single-tenant architecture: Each deployment runs on isolated compute, avoiding noisy neighbor effects and supporting strict data segregation.
  • Unified voice AI stack: Run speech-to-text, text-to-speech, and speech-to-speech workloads in a single runtime with consistent API behavior.
  • Multi-cluster design: Separates real-time, pre-recorded, and agent workloads onto specialized clusters to maximize performance, ensure high availability, and enable strict workload isolation.
  • Region-specific infrastructure: Deploy in your preferred cloud region to meet compliance requirements, enable ultra-low latency, and align with internal policies, including support for country-level deployments.
  • SLA-backed performance: Optional SLAs ensure predictable uptime and latency with defined targets monitored and enforced by Deepgram.

In one modeled scenario, a customer supporting 1,000 concurrent real-time streams would spend approximately $467K USD annually if self-hosting. This includes $250K in DevOps headcount and $98K in infrastructure costs.

Running the same workload on Deepgram Dedicated lowers total OPEX by approximately $98K USD per year. It also reduces engineering overhead and improves deployment reliability through platform-managed SLAs and regional isolation, giving teams more time to focus on higher-impact work.

EU-Hosted API Endpoint: In-Region Inference for European Voice Workloads

Voice AI adoption is accelerating across Europe, driven by demand for real-time applications in finance, public services, retail, and telecommunications. To date, more than two dozen customers and prospects have expressed interest in EU-based infrastructure, highlighting growing demand for in-region processing that meets local performance expectations and regulatory requirements without compromising model quality or flexibility.

To support this, Deepgram is launching early access to api.eu.deepgram.com, a new EU-hosted speech-to-text API endpoint that delivers in-region inference with full feature parity and consistent performance. The EU endpoint is hosted in AWS EU regions, with additional hosting options under consideration.

Key Highlights:

  • Voice data stays within the EU. All processing occurs inside EU-based AWS regions, ensuring no cross-border data transfer.
  • Latency improvements for EU-based users: Localized inference reduces round-trip time for applications serving users in or near the EU.
  • No code changes required: Existing integrations can migrate by updating the base URL, with no other changes needed.
  • Supports GDPR compliance and auditability: The deployment is fully isolated within the EU legal boundary and aligned with regional data protection standards.

This endpoint is well-suited for European ISVs, compliance-focused enterprises, and global teams looking to reduce latency and streamline deployment in the EU.

Why This Matters: A Global-Ready Voice AI Platform

With these additions, Deepgram now supports a range of deployment options, including multi-tenant hosted APIs, fully managed single-tenant deployments, and customer-operated self-hosted infrastructure. This flexibility allows engineering teams to choose the right model based on their application requirements, compliance obligations, and operational preferences. For some, the hosted API provides a fast path to integration. Others may require the regional data residency of the EU endpoint or the isolation and control of a Dedicated deployment. Teams with existing DevOps capacity may opt for self-hosting to align with internal security policies or infrastructure standards.

What differentiates Deepgram is the ability to deliver true flexibility across deployment models. Teams can build and scale voice AI systems using consistent APIs and model performance, while choosing the infrastructure that fits their environment. Looking ahead, the roadmap includes customer VPC deployments, BYOC support, and expanded region availability across Asia-Pacific, EMEA, and LATAM.

Start Building for Your Environment

If you’re building voice applications that require global reach, regulatory alignment, or low-latency performance, now is the time to explore your deployment options. Demand is high, and we’re expanding access selectively:

Deepgram now runs where your business runs. No trade-offs. No overhead. Just voice AI on your terms.

Leave a comment »

Minnesota National Guard Activated To Help With The City Of St. Paul Getting Pwned

Posted in Commentary with tags on July 30, 2025 by itnerd

This is a first for me as I’ve never heard of the National Guard being brought in to help with a cyberattack. But that’s apparently happening in Minnesota:

Minnesota Gov. Tim Walz has activated the state’s National Guard following a cyberattack on the state’s capital, Saint Paul. City officials have not yet disclosed the nature of the cyberattack, but the July 25 breach continues to disrupt city operations and some public services. 

Saint Paul is one of the largest cities in the U.S. with more than 300,000 residents, and is the latest in a list of major cities targeted by hackers and ransomware gangs in recent years.

In a statement on Tuesday, Walz said he authorized the National Guard’s cyber forces at the city’s request to help Saint Paul recover from the cyberattack, saying the “magnitude and complexity” of the breach exceeded the city’s capacity to respond to the incident.

The National Guard will help to “ensure continuity of vital services and the safety and security of Saint Paul residents,” Walz said. 

Saint Paul Mayor Melvin Carter said in a press conference on Tuesday that the city took its government systems offline to contain the intruders, which sparked a citywide Wi-Fi outage.

Carter added that the city does not store much data on its residents, but said there was a risk that data on city employees had been stolen. 

Paul Bischoff, Consumer Privacy Advocate at Comparitech had this to say: 

“This attack has all the signs of ransomware. Ransomware attacks hit local governments all the time, from small rural counties to larger cities like St. Paul. Comparitech researchers have logged 45 confirmed ransomware attacks on US government entities in 2025 to date. These attacks can both steal data and lock down computer systems. City officials must either pay a ransom to restore systems and prevent the release of stolen data, or face extended downtime, data loss, and putting residents at risk of fraud. St. Paul residents should expect a letter from the city in the next few months notifying them if their data was compromised. I don’t fully understand what purpose the National Guard serves here. Most cities and counties go to the FBI for help with data breaches.”

Chris Hauk, Consumer Privacy Champion at Pixel Privacy follows with this:

“Calling in the National Guard for a cyberattack is a new approach, I have not heard of that before. This has all the earmarks of a cyberattack using ransomware, which usually not only holds the affected system’s hostage, but also harvests data from those same systems. Hopefully, the state will be able to determine what data has been stolen and will inform affected citizens of what was harvested. Until then, St. Paul citizens and companies working with the city need to remain on alert for phishing attempts, and other attempts by bad actors attempting to leverage the harvested information.”

Whatever is going on, it’s clearly severe enough to take this action. This illustrates the effects that cyberattacks can have on organizations. Which also illustrates the need to lock environments down so that they don’t get pwned and require this level of response.

1 Comment »

Russian Airline Aeroflot Pwned Leading To The Cancellation Of Flights

Posted in Commentary with tags on July 30, 2025 by itnerd

One of the things about the war in Ukraine is that cyber warfare has become a new front in the war. And this is illustrated by the fact that Russian airline Aeroflot has apparently been pwned by two groups of hackers. Specifically Ukrainian hacker group Silent Crow and Belarusian hacker group the Belarus Cyber-Partisans. According to AP, this has led to the following:

A cyberattack on Russian state-owned flagship carrier Aeroflot caused a mass outage to the company’s computer systems on Monday, Russia’s prosecutor’s office said, forcing the airline to cancel more than 100 flights and delay others.

This was as of Monday. I have to assume that this is still the case today. Here’s where it gets interesting. The groups behind this claim to have been in the network for a year:

Silent Crow claimed it had accessed Aeroflot’s corporate network for a year, copying customer and internal data, including audio recordings of phone calls, data from the company’s own surveillance on employees and other intercepted communications.

“All of these resources are now inaccessible or destroyed and restoring them will possibly require tens of millions of dollars. The damage is strategic,” the channel purporting to be the Silent Crow group wrote on Telegram. There was no way to independently verify its claims.

The same channel also shared screenshots that appeared to show Aeroflot’s internal IT systems, and insinuated that Silent Crow could begin sharing the data it had seized in the coming days.

“The personal data of all Russians who have ever flown with Aeroflot have now also gone on a trip — albeit without luggage and to the same destination,” it said.

The Belarus Cyber-Partisans told The Associated Press that they had hoped to “deliver a crushing blow.” The group has previously claimed responsibility for a number of cyberattacks, and said in April 2024 that they had been able to infiltrate the network of Belarus’ main KGB security agency.

If all of this is true, this is really embarrassing for Russia who is a nation that’s usually associated with attacks like this. The fact that they got pwned on this scale shows that nobody is immune from cyberattacks. Thus this is another object lesson for organizations to make sure that they do everything possible to not be the next victim.

Leave a comment »

Popular Dating Safety App Tea Gets Pwned Big Time

Posted in Commentary with tags on July 30, 2025 by itnerd

A dating safety app called Tea that allows women to do background checks on men and share that with other women has been pwned. As a result, posts, comments, and images that are tied to 1.6 million users are out there in the wild:

At 6:44 AM PST on 7/25, we identified unauthorized access to our systems and immediately launched a full investigation with assistance from external cybersecurity experts to understand the scope and impact of the incident. 

A legacy data storage system was compromised, resulting in unauthorized access to a dataset from prior to February 2024. This dataset includes approximately 72,000 images, including approximately 13,000 selfies and photo identification submitted by users during account verification and approximately 59,000 images publicly viewable in the app from posts, comments and direct messages.

No email addresses or phone numbers were accessed. Only users who signed up before February 2024 were affected.

They later updated this statement with this:

As part of our ongoing investigation, we have recently learned that some direct messages (DMs) were accessed as part of the initial incident. For this reason our DM functionality is down.

To address the issue and out of an abundance of caution, we have taken the affected system offline altogether. At this time, we have found no evidence of access to other parts of our environment.

Please know that we’re committed to keeping you informed as quickly as possible. That said, because this is an active investigation involving external cybersecurity experts and the FBI, there are limits to what we can share—and when. We’ll continue to provide updates as soon as we have confirmed information and are able to do so responsibly.

Our team remains fully engaged in strengthening the Tea App’s security, and we look forward to sharing more about those enhancements soon. In the meantime, we are working to identify any users whose personal information was involved and will be offering free identity protection services to those individuals

Clearly this is bad as it could put women who use the app at risk for identity theft. I will be very interested to see how this is handled by the company. And if they can recover from this incident over the long term.

UPDATE: This has gotten worse for Tea as reports have surfaced that researchers have discovered that user info is now floating around places like the dark web. Anna Collard, SVP Content Strategy and Evangelist at cybersecurity company KnowBe4 Africa, commented:

“The Tea app data breach represents a failure in cloud security and data governance. This incident highlights how misconfigured cloud storage systems can transform what seems like a routine identity verification process, into a privacy breach with potentially lasting consequences. The stolen selfies and drivers licenses could be used by bad actors to ‘confirm’ false identities and commit identity theft. Cybersecurity is not just there for compliance, but to prevent these kinds of incidents, and data protection isn’t just a technical requirement, it’s a fundamental business responsibility that directly impacts user safety.”

Leave a comment »

Anker Powers Up VELD Music Festival 2025

Posted in Commentary with tags on July 30, 2025 by itnerd

This summer, Anker, the global leader in charging technology, is making its VELD debut in electrifying style as the Official Charging Partner of the 2025 VELD Music Festival. From August 1 to 3, Anker will energize Toronto’s premier electronic music event with a multi-brand activation spanning Anker, Soundcore, Anker SOLIX, and eufyMake. This marks Anker Innovations’ first-ever Canadian festival activation, bringing high-speed charging, next-gen personal power solutions, and immersive experiences to over 120,000 festivalgoers throughout the weekend.

As part of the activation, Anker will introduce a network of premium charging stations across general admission areas. The secure, hands-free charging lockers will allow festivalgoers to snap, share, and stream every unforgettable moment without missing a beat.

Anker will display a customized 30′ x 30′ festival footprint with free charging stations, exclusive merch, and immersive brand zones:

  • Anker Live Charged Shop – A pop-up retail space with bestselling portable chargers and cables, and exclusive colours only available at the festival for purchase.
  • Soundcore Soundscape Booth – A relaxing sound zone with immersive audio demos of Soundcore’s latest open-ear and noise-canceling headphones.
  • Anker SOLIX Lounge – A cozy, outdoor-inspired retreat set with comfortable lounge chairs, showcasing Anker SOLIX portable power stations and electric coolers.
  • eufyMake Creation Station – A first-of-its-kind in person look at designs and creations made by the world’s first eufyMake E1 UV Printer, where fans can see live 3D UV texture prints on cases and accessories.

In addition to the zones, festivalgoers can take home lasting memories via the 360° Festival Photo Booth, and the Spin-the-Wheel Station for a chance to win exclusive merch, gadgets, and Anker discounts.

Anker’s presence at VELD underscores a growing commitment to environmental, social and governance (ESG) initiatives, with clean energy innovation playing a leading role. Using energy-efficient technologies across the festival grounds, Anker is helping to set a new standard for sustainable support at large-scale events. Whether powering tech or enhancing comfort, Anker’s renewable energy solutions are designed to elevate the fan experience while reducing environmental impact — all in alignment with its mission to Recharge the Future.

For more information on Anker’s festival activation, giveaway, and product lineup, visit www.anker.com/ca/pages/canada-music-festival and follow @ankerofficial on TikTok and Instagram for live updates throughout the weekend.

Leave a comment »

OpenText and TELUS partner to deliver Canadian sovereign AI-powered solutions for government and business  

Posted in Commentary with tags , on July 30, 2025 by itnerd

OpenText and TELUS today announced the launch of the OpenText and TELUS Canadian Sovereign Cloud. This offering empowers Canadian organizations and government agencies with access to enterprise-grade cloud computing and AI capabilities while ensuring complete data sovereignty — with every application, dataset, computation and network operation remaining securely within Canadian borders. 

This strategic partnership brings together Canadian infrastructure, AI innovation and trusted advisory services to deliver a truly sovereign, secure and scalable cloud AI platform designed specifically for Canadian enterprises and government institutions. Today, OpenText already serves 1,600 Canadian institutions with nearly a thousand organizations actively using and benefiting from AI-powered applications in the cloud. Now, Canadian customers have access to a fully sovereign cloud and AI environment that meets the highest standards of data residency, security and compliance. 

A Canadian-first cloud and AI platform

The OpenText and TELUS Canadian Sovereign Cloud is a purpose-built, enterprise-ready solution that runs entirely within TELUS’ Canadian data centers, including highly secure facilities in Rimouski, Quebec, and Kamloops, British Columbia. It delivers high-performance AI computing capabilities within a secure environment, providing essential sovereign AI compute services for organizations requiring both advanced AI workloads and absolute compliance with Canadian security standards and privacy regulations. 

A commercially proven solution

OpenText and TELUS, both early signatories of the Government of Canada’s voluntary AI code of conduct, have long-standing track records of serving Canadian governments, businesses and institutions. With this new offering, OpenText and TELUS are extending their capabilities to meet the evolving needs of our clients.

In addition, OpenText’s family of Aviator AI products will leverage TELUS’ pioneering AI Factory to offer a sovereign configuration hosted entirely in Canada, enabling customers to realize the benefits of AI-enabled search and summarize while keeping data secure within Canadian borders.  

OpenText’s private cloud solutions are available through existing procurement vehicles and have undergone rigorous technical evaluation, making them ready for deployment today.

Supporting Canadian innovation and growth

The launch represents OpenText’s and TELUS’ broader commitment to Canadian innovation. The OpenText and TELUS Canadian Sovereign Cloud offers a uniquely Canadian solution that combines cloud agility with sovereign assurance, demonstrating how Canadian organizations can compete globally while maintaining complete control over their most valuable data.  

Getting started with Sovereign AI

The OpenText and TELUS Canadian Sovereign Cloud will be available starting September 2025 for commercial and government customers across Canada. For more information or to schedule a consultation, visit opentext.com/sovereign-ai-cloud and telus.com/aifactory.

Leave a comment »

Retab raises $3.5M and launches most powerful document AI platform on the market

Posted in Commentary with tags on July 30, 2025 by itnerd

AI agents are poised to change the world, but they can’t read the documents that run it. Retab, a San Francisco-based startup founded by engineers frustrated by the broken state of document AI, is fixing that. Today, the company announces $3.5 million in pre-seed funding and the launch of its platform.

The round was backed by leading early-stage funds including VentureFriends, Kima Ventures, and K5 Global, alongside Eric Schmidt (via StemAI), Olivier Pomel (CEO, Datadog), and Florian Douetteau (CEO, Dataiku). The new capital will support platform development and community growth, as the company scales its infrastructure to meet rising demand from vertical AI startups and internal innovation teams alike.

Retab is a developer platform and SDK that redefines everything about document processing in the age of large language models. Developers define the schema of the data they need; Retab handles the rest – from dataset labeling and evaluations, to automated prompt engineering & model selection. 

Louis de Benoist and his co-founders cut their teeth building internal automation tools for document-heavy workflows in logistics. Over time, they realized their true value wasn’t in the output, but the orchestration layer they’d built to make the models work. That tooling became the foundation of Retab – now used by dozens of companies to extract structured data from messy, real-world inputs.

Retab is not another large language model. It’s the essential intelligence layer that makes the world’s most powerful models—from providers like OpenAI, Google, and Anthropic—usable for critical workflows. Developers define the data they need, and Retab’s platform manages the entire lifecycle to ensure verifiable accuracy.

The platform delivers guaranteed performance through a system of intelligent checks and balances:

Self-Optimizing Schemas: An AI agent automatically tests and refines instructions based on a user’s documents, maximizing accuracy before the system ever goes live.

Intelligent Model Routing: The platform is model-agnostic. It automatically benchmarks and routes each task to the best-performing model for the job, whether the priority is cost, speed, or accuracy. This can make processes up to 100x cheaper than other solutions.

Guided Reasoning & k-LLM Consensus: Retab forces models to “think” step-by-step and uses a consensus mechanism among multiple models to quantify uncertainty, acting as a powerful safety net to ensure trustworthy results.

Customers across logistics, finance, and healthcare are already seeing results. A major trucking company used Retab and found the smallest, fastest model configuration that could meet their 99% accuracy threshold, dramatically lowering operational costs. A financial services firm uses Retab to extract specific quantitative metrics and qualitative risk factors from 200-page quarterly reports – a task that previously took a team of analysts days to complete. Others are automating claims processing, medical records, identity verification, and onboarding with minimal setup.

Looking ahead, Retab is expanding its platform to apply the same reliable extraction methods to websites and is launching integrations with automation platforms like n8n, Zapier, and Dify. 

Retab is also building toward its long-term vision: to serve as the intelligent middleware layer between the world’s unstructured data and the AI agents that need to understand them. Whether it’s parsing a loan file, a contract, or a customs manifest, Retab makes unstructured data usable, safe, and programmable.

With just ten employees and a fast-growing developer base, Retab is already positioning itself as a foundational layer in the AI infrastructure stack – a tool that doesn’t just show what’s possible, but lets anyone build with it.

Leave a comment »

« Older Entries
Newer Entries »