-
-
AI-driven DevOps & Security Engineer, Automate, Secure, Deploy.
Inspiration
Modern DevOps is powerful but overwhelming. Developers juggle CI/CD pipelines, infrastructure configs, security checks, monitoring, and incident response often across too many tools. We were inspired by a simple question: what if DevOps worked like an autonomous teammate instead of a manual process? Kendra AI was born from the idea of having an AI DevOps engineer that can understand your project, automate decisions, and proactively handle deployment, monitoring, and security without constant human intervention.
What it does
Kendra AI is an autonomous AI DevOps and security assistant that helps developers manage the entire DevOps lifecycle. It analyzes repositories, understands project structure, assists with deployment workflows, monitors system health, and identifies potential security risks. Instead of reacting to failures, Kendra continuously observes, reasons, and suggests (or takes) actions to keep systems stable, secure, and production-ready.
How we built it
We built Kendra AI using a dynamic architecture with a strong focus on AI-driven workflows. The frontend provides a clean interface for interacting with Kendra and viewing DevOps insights. The backend orchestrates AI reasoning, task execution, and system analysis. We integrated LLM-powered agents to analyze repositories, deployment configurations, and system signals. Autonomous logic enables Kendra to reason about next steps instead of relying on hard-coded rules. The system was designed to be modular, making it easy to expand into CI/CD automation, monitoring, and security scanning.
Challenges we ran into
One major challenge was balancing autonomy with control ensuring the AI makes helpful decisions without becoming unpredictable. Another challenge was designing workflows that feel intelligent rather than scripted.
Accomplishments that we're proud of
Successfully combining AI reasoning with real DevOps workflows Creating a product that feels like a teammate, not just a dashboard Designing a scalable foundation that can evolve into a full DevOps platform
What we learned
We learned that autonomy in DevOps isn’t just about automation it’s about context, reasoning, and trust. We also gained deeper insight into designing AI systems that assist developers instead of overwhelming them. Most importantly, we learned how powerful agent-based AI can be when applied to real engineering problems.
What's next for Kendra AI
Next, we plan to add: Full CI/CD pipeline automation Security vulnerability scanning and remediation suggestions Incident detection and self-healing actions Team collaboration features (alerts via Slack/email) Deeper integrations with cloud providers and Git platforms
Log in or sign up for Devpost to join the conversation.