Building offline-first applications and real-time AI vision systems.
Computer Science undergraduate engineering high-throughput FastAPI backends, sub-15ms SQLite caching engines in Flutter, and multi-camera vehicle tracking in PyTorch.
I BUILD DIGITAL SYSTEMS THAT TURN COMPLEXITY INTO CLARITY.
FLAGSHIP PROJECTS
VIT-AP NEXUS
An offline-first mobile application providing sub-15ms cached academic records and instant attendance forecasting.
PARKOS
Smart parking management operating system featuring real-time spatial lot grids, dynamic pricing, and automated check-ins.
OMERTÀ
Multi-camera vehicle surveillance and trajectory Re-ID pipeline tracking across disjoint camera networks in real-time.
VIT-AP NEXUS
Offline-first mobile academic super-app designed to eliminate erratic university portal timeouts and provide instant attendance forecasting.
CASE STUDY: ARCHITECTURE & RESULTS
From erratic student portal to sub-15ms native super-appThe campus web portal suffered from frequent session timeouts, slow DOM rendering, and required re-authenticating repeatedly just to check daily timetable and attendance safety margins.
Designed the entire mobile frontend in Flutter with Riverpod state selectors, built a local SQLite caching engine for instant reads, and engineered an asynchronous FastAPI proxy to parse and structure student records.
PARKOS / SMART PARKING OS
Real-time parking slot reservation and automated session lifecycle platform with dynamic per-second billing.
CASE STUDY: SMART SPATIAL IOT
Automating parking allocation and per-minute revenue meteringCommercial parking complexes struggle with manual paper passes, driver congestion during search loops, and inaccurate hourly billing calculations.
Engineered the complete Next.js frontend with live interactive 2D spatial grid layouts, implemented WebSocket state listeners for instant slot locks, and built dynamic per-second billing logic.
OMERTÀ / AI VEHICLE RE-ID
Multi-camera deep metric re-identification pipeline tracking vehicle trajectories across disjoint surveillance feeds.
CASE STUDY: MULTI-CAM RE-ID
Trajectory continuity across disconnected camera anglesVehicles crossing non-overlapping CCTV networks lose identity context during blind spots, camera occlusion, and variable lighting transitions.
Developed the complete computer vision pipeline using YOLOv8 for vehicle localization, DeepSORT for intra-camera tracking, and deep metric 512-dim embedding extractors for cross-camera cosine matching.
SYSTEM TESTBENCHES [05 PLAYGROUNDS]
Speech → Indian Sign Language (ISL)
Speech-to-text tokenization mapped to structured ISL grammar and real-time gesture avatar animations.
Real-Time Neural Edge & Kernel Visualizer
Interactive image processing lab: Sobel convolutions, Laplacian kernels, and Canny edge algorithms.
Distributed LRU Cache & Raft Consensus
Visualizing lock-free memory eviction, consistent hashing ring, and node cluster failover.
Token Bucket High-Throughput Rate Limiter
Interactive simulation of token refill bursts, sliding window counters, and DDoS traffic throttling.
VTOP Headless Session Scraper Engine
Asynchronous session pooling, encrypted payload parsing, and resilient DOM query engine.
SKILLS & ARCHITECTURE
Flutter
Click any skill on the left to see production applications.
Google Cloud Certified
Associate Cloud Engineer
Demonstrated skills in deploying applications, monitoring operations, and managing enterprise cloud infrastructure on Google Cloud Platform.
Google Cloud Gen AI Academy APAC 2026 — Cohort 3
Hands-on engineering in Generative AI architectures, LLM orchestration, multimodal pipelines, and Vertex AI deployments.
CURIOUS ABOUT HOW COMPLEX SYSTEMS ACTUALLY WORK.
I am a Full Stack Developer & Software Engineer based in Nellore, Andhra Pradesh. My focus centers on building high-performance mobile architectures, low-latency data caches, and real-time computer vision pipelines.
From engineering VIT-AP Nexus to eliminate campus portal wait times with local SQLite caching, to architecting ParkOS for real-time WebSocket telemetry and Omertà for AI multi-camera Re-ID — I build systems tested for reliability and measurable real-world performance.
Started B.Tech in Computer Science & Engineering
Commenced B.Tech in Computer Science & Engineering. Built deep foundations in data structures, algorithms, low-level system designs, and modern full-stack application development.
Started VIT-AP Nexus & Deployed ParkOS
Initiated the development of VIT-AP Nexus to solve campus portal bottlenecks with Flutter & Riverpod. Architected and deployed ParkOS (Smart Parking Operating System) with real-time spatial telemetry and dynamic billing.
Completed VIT-AP Nexus & Developed Omertà
Completed and polished VIT-AP Nexus with near-instant local SQLite caching. Engineered Omertà for multi-camera vehicle Re-ID and trajectory association using PyTorch, YOLOv8, and deep metric feature embeddings.
Building High-Throughput Systems & Next-Gen Software
Designing high-scale distributed architectures, real-time computer vision pipelines, and robust mobile systems.
VIT-AP Nexus & Real-Time Computer Vision Pipelines
Distributed Systems, WebGPU & CUDA Kernel Optimization
Multi-Camera Re-Identification & Agentic AI Architectures
Nellore, Andhra Pradesh, India
Let's Connect.
Have a question, an idea, or just want to say hi? Drop an email or connect directly through my socials.