Sentry AI has been selected as the winner of the SG All-Star Pitch Battle! It was chosen from a field of over 1500 across 150 cities.

Progressing from an initial selection round that had 1500+ companies to industry-category semi-finals where Sentry AI competed with 64 others, the competition featured a grand finale where the remaining 4 pitched to VCs, Silicon Valley leaders, and a global audience of around 5,000 people. At this prestigious event, Sentry AI was crowned the Champion!

This accomplishment is a testament to the entire team at Sentry AI and a direct result of the countless hours of dedication and shared, unwavering belief in our mission.

Thanks to SG for hosting a world-class competition. This win is an honor and a welcome recognition of our work and mission to create a platform that sets a new standard for security and safety, using AI to deter crime and secure our communities proactively!

All-Star Pitch Battle Finals – Sentry AI 3-minute pitch

All-Star Pitch Battle Semi-Finals – Sentry AI 1-minute pitch

LinkedIn Post

Selected in the Founder category, UDAY KIRAN CHAKA joins 50 founders, innovators, and media voices driving the future of AI

SUNNYVALE, CA (Oct, 1st 2025) – Sentry AI announced that its Founder & CEO, Uday Kiran Chaka, has been included in the 2025 Mayfield | Divot AI List in the Founder category, celebrating the top 50 emerging leaders shaping the future of AI. See the complete list here.

The Mayfield | Divot AI List, created by Mayfield, a leading early-stage venture capital firm, and Divot, an initiative of the global entrepreneurial community Startup Grind, recognizes innovators, builders, researchers, founders, policy makers, and media voices shaping the AI landscape. 

Honorees were chosen based on four key criteria: the significance of their contributions to the current and future AI landscape, leadership in innovation, commitment to responsible and ethical AI practices, and potential for meaningful impact or influence in the field.

“Mayfield is a people-first fund, and we believe the future of AI will be shaped by those whose early work is already making a meaningful impact,” said Navin Chaddha, Managing Partner at Mayfield. “The Mayfield | Divot AI List was created to recognize rising stars across the ecosystem whose early work is already shaping what comes next. We’re proud to support and celebrate the individuals driving AI forward with purpose, vision, and impact.”

A panel of respected AI leaders and builders shaped this year’s list. Judges included Dr. Fei-Fei Li, Professor at Stanford AI; Emily Chang, Host and Producer at Bloomberg; Sridhar Ramaswamy, CEO of Snowflake; Rahul Vohra, Founder and CEO of Superhuman; Armon Dadgar, Co-founder and CTO of HashiCorp (IBM); Ion Stoica, Co-founder of Anyscale, Databricks and Professor at Berkeley; May Habib, Co-founder and CEO of Writer; Darren Mowry, VP of Google Cloud Global Startups; Logan Kilpatrick, Product at Google DeepMind; Dheeraj Pandey, Co-founder and CEO of DevRev and Nutanix; Molly O’Shea, Founder of Sourcery; Howard G. Wright, VP of Startup Ecosystem at NVIDIA; Asha Sharma, President, CoreAI Product at Microsoft; Nicholas Thompson, CEO of The Atlantic; Stefano Ermon, CEO of Inception Labs and Stanford professor; Philip Dawson, Head of AI Policy at Armilla AI; Sam Loesche, Head of Policy at Waabi; and Tanya Dua, Senior Tech Editor at LinkedIn.

“Startup Grind exists to give entrepreneurs a platform and a community. By joining forces with Mayfield on the Divot AI List, we’re amplifying the voices of the next generation of AI leaders,” said Derek Andersen, Co-Founder and CEO of Startup Grind. “These are the people writing the next chapter of AI, and we’re honored to celebrate their work and bring them together in San Francisco this fall.”

Together with Startup Grind’s global network, the Mayfield | Divot AI List offers honorees instant visibility to over five million founders, investors, and entrepreneurs worldwide and exclusive access to the rooms building AI’s future. 

Honorees will gather at the historic Golden Gate Club in San Francisco on November 6, 2025, for an invite-only celebration alongside judges, investors, and industry pioneers. The event will feature President & CEO of Y Combinator Garry Tan and a day of meaningful content and connection with peers, press, and industry leaders.

About Uday Kiran Chaka

Uday Kiran Chaka, the Founder & CEO of Sentry AI, is on a mission to build a crime-free society using AI-based security solutions. Uday has more than 25 years of experience in the technology industry, holds 4 patents for innovative computer vision AI applications, and has presented at many industry conferences. Earlier in his career, Uday advised Fortune-100 companies as a strategy consultant at Pricewaterhouse Coopers (PwC). His formal education includes an MBA from Duke University and a B.Tech. from the Indian Institute of Technology.

About Mayfield

Mayfield partners with founders from day zero who see limitless possibilities where others see constraints. Drawing on our 55-year legacy of company building and people-first philosophy, we bring founder-to-founder expertise because we’ve built and scaled companies ourselves. We’ve celebrated over 120 IPOs and 225+ acquisitions, including breakthrough companies like HashiCorp, Lyft, Mammoth BioSciences, and Poshmark. From our Silicon Valley home, with over $3 billion in assets under management, Mayfield is more excited than ever to back founders building the AI future. Join us at mayfield.com.

About Startup Grind / Divot

Startup Grind is the leading global startup community, supporting early-stage founders from bootstrap to Series A and beyond. With members across 150 countries, Startup Grind helps entrepreneurs build companies through events, education, and community.

Divot is the narrative-forward arm of Startup Grind—a community for those making their mark on the world. What began as a podcast has grown into a platform spotlighting legendary leaders and the next generation of rising stars, through storytelling, recognition, and connection.

Media Contact:

Hien Nguyen
[email protected]

Introduction

Securing facilities like office buildings, server rooms, and labs is of paramount importance for enterprises. Enterprise physical security systems typically consist of two independent subsystems: badge-based access control to restrict entry to authorized personnel and security camera systems for perimeter security and monitoring. This article explores the necessity and potential for integrating these two systems to enhance perimeter security and simplify management.

Badge-driven access control

The majority of enterprises and big organizations rely on badge-based access control systems to safeguard their facilities, data centers, and restricted zones. These systems authenticate individuals using personalized ID badges, enabling secure, policy-driven entry management across the organization. Beyond simply restricting access, they offer valuable data for compliance, risk management, and operational visibility. These are usually integrated with many other systems, like Human Resources, Incident Response, and Operations. 

These access control systems are usually centrally managed with software applications like OnGuard, which are robust and powerful. In addition to integration with card readers, they also integrate seamlessly with other systems to provide capabilities, such as

• Dynamic access management based on HR system integration
• Visitor management, temporary access
• Alerts and alarms related to physical entities, e.g., Door unlocked
• Restricted and policy-driven entry management, E.g, main door vs data center entrance access

Camera-based surveillance & monitoring

Camera-based surveillance systems provide real-time monitoring, event detection, and post-event analysis across the facilities. Most of the modern Camera systems and Network Video Recorders (NVRs)  are IP-based and offer features like high-res video, remote access, and recording. These can be integrated with intelligent systems to provide features such as intrusion detection & recognition. These camera systems can also provide audio feeds in addition to the video streams. 

Limitations of having independent systems

While both these systems are very powerful and provide a lot of capabilities, there are certain limitations as well. Access events lack immediate visual verification, making it harder to detect misuse. Without integration, correlating badge activity with corresponding video footage becomes a manual, time-consuming process. This silo’ed approach reduces operational efficiency, delays threat response, and limits the ability to generate actionable insights from combined data. Some of the real examples of challenges faced:

• How to verify that the access card is used by the intended user? 
• Is someone tailgating with an employee?
• Is someone loitering outside the main door to get an opportunity to enter? Or, is it an employee speaking on his cellphone?
• Is the door being held open as a courtesy by an employee to colleagues, or is it left open by mistake?

These are some of the many challenging scenarios that are possible due to a lack of bridging the visual information with access control and physical events. 

Integration using a unified correlation platform

Correlating access control system events with surveillance events can unlock a new level of security intelligence. A unified security platform with access to both pieces of information can significantly enhance situational awareness, operational efficiency, and response capabilities. 

Every badge swipe or access event can be automatically linked to corresponding images from the pre-mapped security cameras, enabling real-time visual verification and faster investigation of anomalies such as unauthorized entry, tailgating, or forced access.

It not only reduces manual effort but also improves incident response time. For enterprises, this convergence can provide situational awareness, enhance operational efficiency in responding to threats. 

Artificial Intelligence-based Smart Guarding has become very effective in providing actionable intelligence and credible deterrence against intrusion and break-ins. These systems, which mostly process visual information (image/video streams), can become truly multimodal when they incorporate data streams such as access control events into a unified decision making framework. By ingesting access control alerts (e.g., badge swipes, denied entries, door held open), the AI engine can correlate these identity-driven events with real-time video analytics, behavioral detection models. 

Fig 1: Integration of Access & Security systems

Sentry AI – OnGuard Integration

OnGuard, a widely used enterprise-grade access control platform, not only provides robust access management capabilities but also supports seamless integration with certified third-party systems via event and alarm subscriptions. Sentry AI is now a certified OnGuard partner and can be configured to receive real-time access events directly from an OnGuard installation. You can read more about the integration here.  

Sentry AI’s cloud-based, scalable Smart Guard platform processes security events through a secure AI pipeline, enabling intelligent threat detection and proactive monitoring. With native integrations across a wide range of security camera systems & Video Management Systems (VMS) such as Milestone XProtect and Eagle Eye Networks, Sentry AI delivers a comprehensive suite of monitoring and deterrence capabilities designed for Global Security Operations Centers (GSOCs) and on-site security teams. The deterrence capabilities are provided by integrating the output systems, such as speakers, sirens, and alarm lights.

For OnGuard integration, Sentry provides a lightweight event subscriber that can be deployed either on-premises or in the cloud, as long as it has IP connectivity to the OnGuard server. This subscriber can be configured to listen for standard OnGuard events such as “Access Granted,” “Access Denied,” “Door Held Open,” and more.

Sentry’s user interface also provides a way to configure a mapping between access control devices (e.g., card readers) and associated security cameras. This enables Smart Guard systems to correlate access events with camera events, providing contextual alerts such as tailgating or unauthorized access to security operators. Users receive access-related alerts enhanced with visual elements such as live views and person or vehicle recognition, providing better and comprehensive situational awareness.

For example, the “Main hallway” camera can be mapped “Main entrance” card reader, which enables the user to view the video streams and image alerts from the Main hallway camera every time an access event happens on the Main entrance. 

Fig 2: Sentry-Onguard integration with Sentry being the unification layer

Sentry AI serves as an intelligent unification layer, offering powerful capabilities for proactive monitoring, threat deterrence, and actionable insights from access control and security systems. Most importantly, it provides a holistic view of on-premise activities, enhancing the situational awareness and operational efficiency of security teams tasked with protecting the facility.

Summary

Integrating badge-based access control with camera-based surveillance is crucial for enhancing enterprise physical security. While both systems are powerful independently, their siloed operation creates vulnerabilities and inefficiencies. A unified platform, like Sentry AI’s Smart Guard, bridges this gap by correlating access events with real-time video & image alerts. This integration enables visual verification of badge usage, detection of tailgating or loitering, and immediate alerts for anomalies. By leveraging AI, such integrated systems provide actionable intelligence, proactive threat deterrence, and a holistic view of on-premise activities, significantly improving situational awareness and operational efficiency for security teams.
If you want to try out our state-of-the-art security solutions, please reach out to us here!

According to a Sunnyvale Police Department press release, a group of suspects rammed a vehicle into a local jewelry store and made off with stolen items before officers could arrive. This was not in the night or early morning. This was broad daylight. In the middle of the afternoon.

Would a Security Guard Have Helped?

It’s tempting to think that posting a guard might have stopped this—but in reality, it might have escalated the situation. When faced with a fast-moving group determined to smash, grab, and flee, a single guard is unlikely to intervene safely. Worse, they may become a target themselves.

What Actually Works: Layered Security

To prevent incidents like this, businesses need more than a warm body at the door. They need strategic, layered defenses:

• Environmental Design (CPTED): Bollards, hardened storefronts, and better visibility discourage break-ins
• AI-Monitored Security Cameras: Real-time monitoring catches loitering and unusual activity before an incident
• Speakers & Remote Guard Intervention: Talk-down systems that confront intruders before damage occurs
• Rapid Escalation Protocols: Clear video evidence, suspect descriptions, and license plate info delivered to police immediately

Every Second Counts

When criminals are in and out in under 90 seconds, there’s no time to call for help later. The security system must detect, deter, and escalate as it’s happening. That’s where Sentry AI’s SmartGuard solution comes in—our platform uses AI to detect suspicious behavior, trigger live intervention, and alert authorities with usable intel.

We All Pay the Price When They Get Away

Every time a crew escapes, they learn. They repeat. They hit again—often in a different city. Jewelry stores, cannabis shops, mobile phone retailers… no one is safe without the right preparation.

So What Should We Do as a Society?

• Should we require bollards or barricades outside high-risk stores?
• Should real-time AI security monitoring be mandatory in sensitive zones?
• How do we stop robbers from feeling like they can act with impunity?

Until we answer these questions, we’ll keep seeing videos of cars slamming into storefronts while the public shakes its head and asks, “How did they get away with it again?”

It’s time for a new approach to crime prevention. Not just reactive, but predictive. Not just guards—but SmartGuards.

Learn more about how Sentry AI can protect your business.

One stolen key can compromise an entire neighborhood.

A Smarter Lock: Two-Factor Protection for Your Mailboxes

At Sentry AI, we knew the real solution wasn’t just better locks—it was smarter access control.

That’s why we built MailGuard, a patent-pending security solution that combines:

• The USPS master key, and
• A secondary combination lock that only the USPS delivery person or property manager knows

Just like two-factor authentication protects your email, MailGuard protects your mail.

Real-World Success: Mail Theft Eliminated

We’ve already deployed MailGuard in three communities. The result?

Zero mail theft.

Here’s why it works:

• Thieves can’t act fast – The coded lock adds a time delay
• Wrong codes waste their time – Giving authorities time to respond
• Fake codes protect carriers – If held up, USPS personnel can give a dummy code without immediate consequence
• Visible deterrent – The upgraded box signals: this community is protected

Add Smart Surveillance: Sentry AI’s SmartGuard

MailGuard works even better when paired with our SmartGuard AI monitoring system:

• AI-powered SmartGuards detect loitering, tampering, and unauthorized access
• Loudspeaker talk-downs stop crime before it starts
• Instant alerts to property managers or law enforcement
• Thieves walk away after spotting the upgrade

In one community, thieves were caught on camera approaching a MailGuard-equipped box—only to turn around and never return.

Word is Spreading

We didn’t expect the response to be this fast.

Mail carriers, property managers, and HOA board members are already sharing MailGuard by word of mouth. Some USPS delivery personnel have told us it’s the first time they feel safe delivering mail in certain neighborhoods.

Now, we’re getting inbound proposal requests from communities all over the U.S.

Arrow Key Theft Ends Here

MailGuard is more than a lock—it’s a system:

• Deterrence
• Delay
• Defense
• Documentation

And most importantly: peace of mind.

Want to protect your community from mail theft?
Get in touch to schedule a site evaluation and learn more about Sentry AI MailGuard + SmartGuard.

That key exists. It’s called the Arrow Key, and it’s at the center of a national mail theft epidemic.

What’s Happening?

Arrow Keys are master keys issued by the U.S. Postal Service (USPS) to mail carriers and supervisors to access cluster mailboxes and blue USPS drop boxes.

But in recent years, these keys have become prime targets for criminals.

According to CBS News and USA Today, Arrow Keys are:

• Stolen or sold on the dark web for $1,000 to $7,000
• Used by organized crime rings to steal mail across neighborhoods
• Fueling identity theft, fraud, and check washing at scale

In one case, The Sun reported that a single stolen key led to theft from over 100 homes—without breaking a lock.

The Impact

• Millions of dollars lost every year from mail fraud
• Victims dealing with ruined credit and identity theft
• Postal workers attacked or robbed for their keys
• Public trust in USPS eroding

What Is USPS Doing About It?

USPS has taken steps to combat Arrow Key theft, such as:

• Installing electronic locks in some high-risk areas
• Urging customers to use secure indoor drop locations
• Offering rewards for information
• Increasing enforcement through USPS Inspection Service

Still, there’s no foolproof solution yet. As long as a single key can open hundreds of mailboxes, criminals will try to get their hands on it.

There Has to Be a Smarter Way

At Sentry AI, we believe the solution doesn’t lie in stronger locks—but in smarter protection.

In our next post, we’ll unveil our patent-pending technology that uses Cybersecurity-inspired deterrence, real-time detection, and intelligent alerting to secure communities—even when Arrow Keys fall into the wrong hands.

Mail theft doesn’t just steal envelopes—it steals trust, time, and peace of mind.
Let’s take that back.

Stay tuned for our next post.

Trespassers busted by Sentry AI’s SmartGuard

Hot tub at 4:30 AM? Even behind gates? Trespassers thought no one would see.
But Sentry AI’s SmartGuard was watching — and acted.

The system immediately detected the unauthorized activity, issued live voice warnings, and alerted the police. Trespassers ignored the warnings… until the cops showed up and cleared them out. Lesson learned: Don’t mess with SmartGuard.

This Silicon Valley community had suffered for months:
Mail theft
Break-ins
Trash left behind by late-night trespassers
Costly, ineffective patrols

The HOA Board had tried everything — cameras, locks, guards — but nothing worked. Then they discovered Sentry AI.

The challenge?
No power. No internet.

Our solution:
Solar-powered
Battery-backed
Cellular-connected
Wireless-bridged system

Deployed, tuned, and ready. Just days later — the first big win.
Residents woke up to peace of mind, finally feeling secure.

We had the opportunity to talk to almost 350+ companies in the event, of which more than 100 were interested in using our solutions directly or through their partners. Sentry provides solution through images by using Computer Vision and Machine Learning.

applications these companies were looking for some of them were in defense which included constantly monitoring the areas of interest and sending alerts if there are signs of specific objects. Drone surveillance was used to know the location of objects in a certain area, public safety to monitor traffic, number of vehicles and intrusion. Enterprise solutions included drones monitoring Parking Lot, Oil spillage and Mining area.
Along with image processing Sentry wants to combine different drone technology to come up with a full integrated partnered solution. Companies like CAPE use drones (DJI, parrot etc.) to provide the users with more capabilities like selecting the perimeter, max height, min height etc. and wants us to configure the drone based on Artificial Intelligence. There were companies who manufactures cameras for drones to receive video feed directly as a networking protocol. Companies producing synthetic data for drones, doing analytics based on sensor data, working on embedded systems etc. We also came across some companies who provide opportunities for a start up like Sentry to participate in competitions and win big money.

In conclusion, Sentry was seen as the partner whose current technology will make their product more valuable and vastly improve the quality of service that these companies provide to their customers. There were a few competitors for Sentry but most of them were in the data analytics space and the ones (may be 2 or 3) focusing on AI had inferior results.

Overall we had a great learning experience in terms of technology and requirements and the city of Chicago was beautiful.

For the first time, artificial intelligence, or AI, cameras are being tested in the UOG Office of Information Technology through the DOCOMO 5G Open Partner Program, which provides global businesses and organizations with the latest 5G information, test environments, and partner workshops.

As one of the AI systems in the Program, Sentry AI focus on learning over time how to automatically spot suspicious behavior through a CCTV (closed-circuit television) feed, while recognizing and tracking who is present and when.

The University found the new AI technology eventually improve the security of the campus

Sentry AI system overlay on a live CCTV stream to study and differentiate movements. In what is known as deep learning, the Sentry AI has spent several months observing and learning “typical” behaviors in the IT office. The AI captures and flags out-of-the-ordinary situations and sends a message or photo via alerts, either through SMS texts or email, as immediately as the incident happens.

The technology, for example, could differentiate between a person walking to the office during a regular working day versus someone suspicious with a covered face or carrying a weapon or someone who otherwise shouldn’t be in an area.

The primary objective with these technologies is to verify the AI’s ability to operate on DOCOMO PACIFIC’s 5G network versus its 4G network. The lower latency of the 5G network allows cameras to transmit high-quality video to Sentry AI, which then recognizes people and activities with higher accuracy. For example, in addition to recognizing people by face, Sentry AI can process live video to identify people uniquely by their body shape and walking patterns, even when a face is not clearly visible.

For University of Guam, having an AI CCTV system could help identify potential threats around campus in real time, which would allow for quicker response times.

“With these projects, we’re getting the opportunity to explore the capabilities and benefits of the latest AI technology, which may lead us to new ways to improve the security, safety, and function of our campus,” said Manny B. Hechanova, UOG’s interim chief information officer.

Everyone has heard the story of the Boy Who Cried Wolf, but do you know there is a shepherd boy in your home as well? He looks after your home, however… Check out the video and see what happened!