2017 — PRESENT A CAREER IN SECURITY

The long view.

From distributed infrastructure to autonomous AI agents. Nearly a decade of building security into the systems people depend on.

01 / FOUR CHAPTERS. A WIDER VIEW.

DEC 2023 — PRESENT · BENGALURU, INDIA

Greenlight

Staff Security Engineer, Cloud / AI Security

Securing financial infrastructure.
Building for the age of AI.

I lead cloud and AI security across infrastructure and data serving millions of families.

  • Make secure defaults the starting point. Architect patterns across multi-account AWS, Kubernetes, and Databricks-integrated AI/data workflows: workload identity, network segmentation, least privilege, CI/CD policy enforcement, data access, and agent/tool execution boundaries.
  • Build security operations with AI. Built an internal, multi-tier system on AWS Bedrock, deployed with Terraform. Designed scoped tool and Lambda permissions, audit logging, approval workflows, and prompt-injection-aware handling of untrusted enterprise context.
  • Connect architecture to operations. Lead CNAPP/CSPM, vulnerability management, threat detection, and IaC security programs, alongside threat modeling and reviews for cloud-native applications, Kubernetes, data, AI/MCP workflows, and agentic infrastructure.
Systems & integrations

AWS · Kubernetes · IAM · WAF · Terraform · Bedrock · Databricks · Okta · Jira · Confluence · Slack · GitHub · Tines · Snowflake · Datadog

Joined as Senior Cloud Security Engineer in December 2023. Staff since January 2025.

APR 2022 — OCT 2023 · BENGALURU, INDIA

Tenable

Senior Cloud Security Research Engineer

Turn cloud threat research
into product capabilities.

I helped build Tenable Cloud Security from the ground up, connecting research to the security decisions customers make every day.

  • Research the attack paths. Investigated emerging threats across AWS, Azure, and GCP and translated findings into detections, attack-path analysis, and security controls.
  • Improve detection quality. Developed MITRE ATT&CK-based cloud detection rules, improving detection accuracy by 25%.
  • Move checks into development. Built IaC security plugins to identify misconfigurations before deployment, alongside benchmark checks and customer-facing security guidance.
Research & engineering focus

AWS · Azure · GCP · MITRE ATT&CK · Attack-path analysis · Infrastructure as Code · Cloud detections · Security benchmarks

Cloud security research across AWS, Azure, and GCP.

APR 2020 — MAR 2022 · INDIA

Cisco

Software Security Engineer, Supply Chain Security

Build trust into identity
and the software supply chain.

I built cloud-native identity and threat-mitigation capabilities, bringing security tooling into engineering workflows.

  • Strengthen identity boundaries. Built capabilities with SSO, OIDC/OAuth 2.0, PKI, and WAF.
  • Understand the whole architecture. Performed threat modeling across Kubernetes, monoliths, and microservices.
  • Make security reusable. Integrated security testing into development and built tooling adopted by engineering teams, reducing identified vulnerabilities by 40%.
Engineering focus

SSO · OIDC · OAuth 2.0 · PKI · WAF · Kubernetes · Threat modeling · Supply-chain security · Security testing

Security capabilities built for adoption by engineering teams.

JAN 2017 — APR 2020 · BENGALURU, INDIA

Intel Corporation

System Programmer, Security

Start with the systems
underneath everything.

I built security automation for high-compute and distributed infrastructure, supporting security operations for global teams.

  • Automate remediation. Built vulnerability remediation and patching workflows that mitigated 90% of vulnerabilities within 48 hours.
  • Follow incidents to the root. Performed incident response, forensics, and root-cause analysis across distributed environments.
  • Work across infrastructure. Supported the security needs of high-compute systems and global operations teams.
Engineering focus

Security automation · Distributed infrastructure · Vulnerability remediation · Patching · Incident response · Forensics · Root-cause analysis

The first chapter: security engineering close to the infrastructure.

02 / ADVISORY & PRODUCT DESIGN

A practitioner’s perspective.
Earlier in the process.

Helping emerging security products take shape through the experience of building and operating security systems.

MAR 2026 — PRESENTRemote

Rig Security

Security Advisor

Advise an emerging identity security platform focused on human, non-human, and AI-agent identities, identity correlation, access governance, and attack-path visibility.

Identity security
JAN 2026 — PRESENTRemote

AirMDR

Early-Stage Design Partner

Partner with the founding/product team on an AI-powered Managed Detection and Response platform, contributing practitioner perspectives on security operations, threat detection, investigation, and response workflows.

AI-powered security operations

03 / AI SECURITY RESEARCH & OPEN SOURCE

The next boundary
is an agent’s action.

Independent work on deterministic authorization and runtime containment for autonomous AI agents.

01 / ACTION & TRAJECTORY CHECKSBeta

Agent Circuit Breaker ACB

Creator / Maintainer · Python, Linux, AI Agents, MCP

A deterministic runtime safety layer that evaluates agent actions before execution across shell, filesystem, SQL, and MCP/tool operations. Includes trajectory-level controls, destructive-operation detection, secret-to-egress detection, approval workflows, and auditable decisions.

Explore the project Source code
02 / AUTHORIZATION & CONTAINMENTResearch prototype

KECHAA

Developer / Researcher · Python, Linux, AI Agent Security

Kernel-Enforced Containment for Hijacked AI Agents. Binds tool commands to task contracts and single-use signed grants, with approval workflows, replay protection, and signed audit records. An experimental Linux backend uses Landlock, seccomp, namespaces, and cgroups to restrict commands and their descendants.

Explore the source Read the architecture

04 / TECHNICAL EXPERTISE

Across the stack.
Connected in practice.

The interesting security problems live between disciplines. These are the layers I work across.

01AI & agent security

Agent security · MCP · Tool authorization · Runtime safety · Agent trajectories · Capability security · Prompt injection · AI security architecture

02Cloud & infrastructure

AWS · Azure · GCP · Kubernetes · Terraform · CI/CD · CNAPP · CSPM · CWP

03Identity & data

IAM · Workload identity · Non-human identity · Least privilege · Just-in-time access · Data security · DLP

04Engineering & operations

Python · Linux · SQL · Shell · Policy-as-code · Threat modeling · Security automation · DevSecOps

05 / KNOWLEDGE THAT TRAVELS

Built in practice.
Shared with the field.

STANDARDS

Center for Internet Security

Benchmark Co-Author · Apr 2022 — Present · Remote

Co-authored recommendations for CIS Microsoft Azure Foundations Benchmark v2.0.0 and CIS AWS Foundations Benchmark v1.5.0, translating cloud security research into industry-standard guidance.

About CIS Benchmarks

EDUCATION

The foundations

Gujarat Technological University

Bachelor of Engineering, Information Technology

2012 — 2016
IIT Gandhinagar

TEQIP-II Long-Term Exchange, Artificial Intelligence

2014

THE NEXT CHAPTER

Something worth building?
Let’s compare notes.

Security engineering, AI research, advisory work, or a conversation about what comes next.

Get in touch chhatralasagar99@gmail.com