Welcome to Vice City
Senior Full Stack Developer | AI & Computer Vision Researcher

I'm a passionate software engineer with a Master's in Computer Science from Northeastern University. With 9+ years of experience spanning full-stack development, game development, and AI/ML research, I bridge the gap between cutting-edge research and production-ready applications.
My research focuses on Computer Vision, Deep Learning, Medical Image Analysis, and Explainable AI. I've developed systems for skin cancer classification, monocular depth estimation, and multi-modal book detection using state-of-the-art architectures.
Architected and shipped a HIPAA-compliant care-coordination platform on AWS Amplify serving 200+ caregivers and patients, with real-time messaging and shared care-plan editing for collaborative Alzheimer's care. Led the Amplify Gen 1 to Gen 2 migration, cutting deploy time by 40%, and unified 3 fragmented frontends into a TypeScript monorepo with a 40+ component design system. Hardened the platform for HIPAA and mentored 2 junior engineers.
Drove migration of a legacy ASP.NET Web Forms real-estate analytics application to Blazor WebAssembly, cutting initial page load from 4.2s to 2.1s. Designed a Blazor component library of 30+ reusable UI primitives, built a real-time SignalR market-data pipeline serving 500+ concurrent users with sub-200ms latency, and introduced an xUnit + Playwright test suite that raised code coverage from 12% to 68%.
Developed a robotic cashier and self-service kiosk in C# / .NET processing 500+ daily transactions, integrating with POS terminals and payment hardware over TCP/IP and serial protocols. Shipped a Xamarin mobile ordering app and ASP.NET Core / Angular admin dashboard, cutting kitchen ticket time by ~25%, and built a hardware abstraction layer that ran across 3 machine variants.
Built multiplayer networking for a VR smart-city simulation in Unity, supporting 20+ concurrent users with sub-100ms state sync via MLAPI and SteamVR. Implemented spatial audio, hand-tracking, and HUD systems, plus a spectator/desktop client for non-VR participants. Authored an asset pipeline that cut per-scene iteration from 4 hours to 20 minutes.
"Sukhrob is one of the most versatile developers I've worked with. He took a vague product idea and turned it into a polished, production-ready application in just 8 weeks. His ability to handle everything from database design to pixel-perfect UI is genuinely impressive. Already planning our next project together."
"We brought Sukhrob on to rescue a project that was falling behind. He quickly identified the architectural bottlenecks, refactored the core systems, and delivered a stable release two weeks ahead of our revised deadline. Clear communicator, great problem solver, zero drama."
"Finding a developer who understands both game networking and web infrastructure is nearly impossible — Sukhrob is that rare find. He built our entire multiplayer backend and admin dashboard from scratch. The codebase he left us is clean, well-documented, and easy to maintain."
Developed explainable deep learning system for multi-class skin lesion classification using EfficientNet V2 with GradCAM++ visualization. Implemented novel ABCDE criterion analysis for clinical interpretability across all 9 ISIC 2019 categories.
Designed lightweight depth estimation model with ResNet18 encoder and U-Net decoder, achieving 42% fewer parameters than Depth Anything V2, 72% faster inference, and better relative error metrics on NYU Depth V2 benchmark.
Implemented FSRCNN architecture achieving 40x speedup over SRCNN with end-to-end learned upsampling. Evaluated on Set5 (+1.78 dB PSNR) and Set14 (+1.26 dB PSNR) with comprehensive ablation studies.


Community platform for CS2 servers serving 30K+ registered users and 1K+ daily active players with social features, in-game admin/ban/report system, Stripe-powered shop, and real-time server integration.


HIPAA-compliant platform connecting to hospital PACS via DICOM for AI-powered analysis on X-ray, CT, and MRI studies with OHIF Viewer integration.

Online chess platform with PvP and AI opponents, supporting rated and unrated matches with real-time WebSocket communication.
End-to-end book detection from bookshelf images using YOLO segmentation and Moondream2 vision-language model for title/author extraction.
Northeastern University, Boston, MA
January 2024 - May 2026
GPA: 4.0/4.0Suffolk University, Boston, MA
September 2021 - May 2023
Cum Laude | GPA: 3.5/4.0INTI International College, Malaysia
September 2019 - July 2021
TUIT, Uzbekistan
September 2017 - June 2019
GitHub
github.com/suxrobgmTelegram
@suxrobgm