About
Started in 2018 with a Diploma in Computer Engineering — when curiosity felt sufficient and a full degree felt optional. Got the B.E. anyway in 2024 (CGPA 8.88), but the real education happened between commits.
Since 2023 I've been building at Noovosoft Technologies — first as an intern, now full-time. I shipped a recurring billing engine that cut revenue failures by 30%, migrated a weather data provider improving accuracy by 25%, and slashed CI/CD build times by 45% by adopting uv across our Python pipelines.
Outside work I built Vartalaap (a WebRTC SFU video platform with real-time AI captions), Servio (a Go-based Linux service manager), and django-silky — a production-quality fork of django-silk with dark mode, D3 analytics dashboards, and N+1 query detection, built in a single day.
I care about distributed systems that hold up under pressure, open source tools developers actually reach for, and code that doesn't need a 40-minute explanation. When I'm not shipping, I'm playing cricket or reading about something I probably can't use at work yet.
Technical Expertise
Resilient Distributed Systems
Background tasks, async flows, and event-based architectures (WebRTC/WebSockets).
Payments & Subscriptions
Robust processing with retry scheduling and user-facing status reporting.
DevOps & Deployments
Systemd services, Nginx configuration, and self-hosted tooling.
Frontend & Mobile UX
Building Angular, React, & React Native interfaces with smooth, polished UX.
Database Performance
Optimizing query patterns and relationships for high scale.
Mobile Offline-First Sync
React Native apps with MobX state, background sync, and Expo SDK 54.
Technical Skills
●●● expert · ●● proficient · ● familiar
Experience
Noovosoft Technologies LLP
Software Developer
Sep 2024 — Present
Software Developer Intern
Aug 2023 — Sep 2024
Architected the recurring billing engine for Farmdesk SaaS, boosting revenue reliability by ~30% via robust retry logic and Mollie integration. Led the migration from DTN to OpenMeteo, improving data accuracy by 25%. Optimized CI/CD pipelines with `uv` to cut build times by 45% and resolved critical race conditions in government animal database integrations.
PythonDjangoAngularTypeScriptCeleryRedisPostgreSQLDockerWebRTC
Featured Projects
Technical Highlight
Vartalaap
Scalable SFU Video Conferencing PlatformArchitected a scalable Selective Forwarding Unit (SFU) video architecture to support high-quality multi-party calls with minimal bandwidth. Unlike P2P mesh, this central media router efficiently manages streams, decoding once and forwarding to participants. Integrated OpenAI Whisper (C++ optimized) for real-time AI captioning and handled complex signaling via WebSockets in Django.
Engineering Decision
Why SFU over P2P mesh? In a mesh, each peer uploads to every other peer — O(n²) bandwidth. SFU routes raw streams through one server — O(n). For any call with more than 3 participants, SFU is the only architecture that scales.
- Next.js
- Django
- WebRTC (SFU)
- Fast-Whisper
- Redis
- PostgreSQL
- Docker
- servio — journald[●] api-gateway.serviceActive: active (running)D-Bus: connected to systemdLogs: streaming via journald▋CPUMEM
Technical Highlight
Servio
GUI-Based Systemd Service ManagerBuilt a visually rich Web GUI for orchestrating Linux servers, replacing command-line fatigue with a modern dashboard. Directly interfaces with Systemd via D-Bus to start/stop services, view real-time journald logs, and manage environment variables. Includes a Git-integrated deployment pipeline that automatically builds and reloads services on push.
Engineering Decision
Why Go over Node.js or Python? Needed a single self-contained binary with no runtime dependencies, direct D-Bus bindings for Systemd, and native goroutines for concurrent log streaming across multiple services. Go's stdlib handles all of this without a single external dependency.
- Go (Golang)
- Linux Systemd
- WebSockets
- React UI
- SQLite
- D-Bus
- django-silkydark/api/users/240ms/api/posts/38ms/api/tags/12msflame · /api/users/ORMN+1SerNet⚠ N+1 detected · /api/users/
Technical Highlight
django-silkyBuilt in 1 day
Production-Quality Fork of django-silkForked the popular django-silk profiling library and shipped a fully modernized version: persistent dark/light theming, inline collapsible filter bar, D3.js analytics dashboards, N+1 query detection with endpoint attribution, and self-hosted icons (zero CDN dependencies). Drop-in replacement — no new migrations required.
Engineering Decision
Why fork instead of contributing upstream? The changes required architectural rewrites — CSS variables for theming, D3 for analytics, self-hosted assets. Getting that through upstream review would take weeks. Forking let me ship in one day and write about it.
- Python
- Django
- D3.js
- CSS Custom Properties
- PostgreSQL
Recent Blog Posts
django-silky: A Modern Fork of django-silk with Dark Mode, D3 Charts, and N+1 Detection
Feb 18, 2026•4 min readIf you've ever profiled a Django app, you've probably used django-silk. It's invaluable — every HTTP...
#django#python#webdev#opensource
Modern Django & Django REST Framework Complete Roadmap
Aug 12, 2025•14 min read•1 reactionsGetting Started Installation & Setup Setup Django with Docker...
#django#drf#djangorestframework#webdev