Hey, I am Aayan

I am a software engineer passionate about bridging the gap between systems programming and artificial intelligence — building efficient, scalable, and intelligent systems from the ground up. My work spans backend architecture, frontend engineering, and deep learning research.

I currently work as an SDE Intern at Aftershoot, where I design and build production Rust crates from the ground up — architecting clean abstractions and type-safe APIs. I authored a standalone image-metadata library handling EXIF and XMP read/write across JPEG, PNG, JXL, HEIF, and RAW formats, and wrote declarative procedural macros that generate strongly-typed structs and builders, integrating a C library through safe Rust FFI.

I was selected for Google Summer of Code 2026 with MoFA, where I'm architecting the core of an open-source Rust engine that orchestrates multimodal AI models (LLM/TTS/ASR) across local and cloud backends — with deterministic sub-millisecond routing, memory admission, and automatic failover — shipped as a versioned HTTP API and a native Rust SDK with UniFFI Python bindings.

Previously, I built a production real-time AI voice agent using Deepgram's STT and TTS APIs (with FastAPI and WebSockets) serving 1,000+ active users, cutting voice infrastructure costs by 65% and end-to-end latency by 40%. At the Software Development Cell, GGSIPU, I lead backend development for university-scale platforms used by thousands of students — designing scalable APIs, managing databases, and deploying secure infrastructure on VPS environments.

Beyond web systems, I actively explore the intersection of machine learning, deep learning, and natural language processing. I experiment with large language models, reinforcement learning, and multimodal architectures, constantly pushing my understanding of how intelligence can emerge from computation.

I'm deeply passionate about Rust and systems programming — fascinated by its blend of performance and safety. I use it to understand how low-level design impacts scalability, performance, and the developer experience itself.

I love dissecting complex ideas — whether it's a neural architecture or a compiler optimization — and translating them into real, working systems. I continuously share my insights and learning journeys across my website and socials, driven by the belief that engineering is a lifelong process of curiosity and creation.