"81% of web-based applications are compromised as a result of stolen credentials or rogue users"
(Source: 2017 Verizon DBIR)



The Yaxa Analysis Engine utilizes unique reinforced ML to monitor user traffic accessing critical web applications building a unique profile for each user’s behavior. It extracts, analyzes, and builds behavioral profiles from live data packets and captures identity information from existing IAM servers. We prevent loss of valuable data by following a “trust but verify approach,” with minimal IT intervention. This is done via a robust four-step process:

1. Behavioral Fingerprinting

YAXA enables user and application specific access controls. User behavioral traits such as; time and duration of access, speed, geolocation, user agent, client device, response types, application zones accessed, sequence of operations, upload and download behavior to build a deep behavioral model of all users accessing your web-based applications. Yaxa’s behavioral fingerprinting is the first process in protecting your valuable assets. No more fine tuning of static rules.

2. Anomaly Detection

Our dynamic behavioral fingerprinting process automatically establishes user baseline from which YAXA detect real-time anomalies as the users or devices access the assets under protection. Yaxa employs an ensemble of advanced machine learning algorithms developed in collaboration with a top technology university. The ensemble method boosts accuracy of anomaly detection maximizing coverage of user or devices behavior and minimizing false alerts.

3. Blocking Attacks

Upon detecting an anomalous session, Yaxa’s automated Incident response mechanism triggers an authentication request to vet the user in real-time. Flexible policy allows administrators to customize profiles based on perceived threat, risk sensitivity, and user privileges. YAXA can distribute an authentication request to an individual user, manager or multiple persons based on company policy.

4. Adaptive Learning

Yaxa constantly incorporates user feedback into behavioral models in order to adapt to changing user behavior and risk patterns. With such unsupervised and adaptive learning, Yaxa is able to reduce future alerts while responding to ever changing threat landscape.

Product features


Our Team


Kalpesh Sheth

Co-founder and CEO


Ramesh Gupta

Co-founder and CTO



Human attack surface by 2021

Average security alerts per day

Percent of network traffic is encrypted

contact us for demo

Contact Us

Yaxa can provide a demo for your custom/internal application or SaaS applications like SalesForce, Office365, DropBox, Gmail, Box and other web based applications.

  • +1 (978) 727 4811
  • 300 Baker Ave, Suite 300
  • Concord
  • MA 01742

To request a demo or to know more about Yaxa solution, please reach out to us: