About

Hi! I’m Kalyan Cheerla — also known as kc, gh0st, or kite online.

I’m a Research Engineer working at the intersection of AI, security, and systems — building tools that make machine learning trustworthy and private by design. My work blends research rigor with a builder mindset: from cryptographic ML pipelines to secure, self-hosted infrastructure.

What I’m currently exploring

• Exploring the shift to probabilistic systems through LLMs, and how they enhance deterministic rule-based logic (Rule-based Logic + LLMs = magic ✨).
• Trying to build a custom Garbled Circuits framework for secure computation.
• Blogging about tools, workflows, and technical learnings I wish I had found earlier.

Recent work & projects

• 🧠 FHE vs GC Paper: Cheerla, K., Ben Othmane, L., & Morozov, K. (2025). Comparison of Fully Homomorphic Encryption and Garbled Circuit Techniques in Privacy-Preserving Machine Learning Inference [Preprint]. arXiv. https://arxiv.org/abs/2510.07457

• 🛠️ MicroSvcsThreatsMapper — built a Threats Manager Studio extension to automate threat model generation for Kubernetes microservices aiding PhD research. Code repository with Video

• 🛠️ VoxPilot — client-side, privacy-preserving AI assistant using Gemini Nano for on-device voice interaction. VoxPilot Blog

Previously

• Built and deployed MedVoyage, a full-stack healthcare web app using Django, MySQL, Docker, and k3s.
• Served as Teaching Assistant for Secure Software Systems over four semesters, creating hands-on cybersecurity labs.
• At Wipro (Mastercard), wrote >1 million lines of C/C++ for global payment authorization systems — and learned what real-time reliability actually means.

Beyond work

I maintain a self-hosted homelab, experiment with AI agents and privacy-first automation, and document technical deep dives on this site. I love bridging gaps between research prototypes and production-grade systems — making cutting-edge ideas actually deployable.

Open to Collaborations

I’m currently open to research collaborations, industry projects, or full-time opportunities in AI Security, Privacy-Preserving ML, or Applied Systems Research.

If you’re working on something aligned — whether it’s academic, open-source, or experimental — I’d love to connect. Reach me at [email protected] or via below socials.

Elsewhere

• Resume: View PDF
• Google Scholar: scholar.google.com/citations?user=wsmUsV0AAAAJ&hl=en
• GitHub: github.com/kalyancheerla
• X (Twitter): x.com/kalyanch05
• LinkedIn: linkedin.com/in/kalyancheerla