- ⚙️ Backend engineer with 6 months of production experience at a London-registered B2B SaaS startup (SvaraAI)
- 🔧 Built systems from scratch — 14K req/sec HTTP server in C, 18.8K req/sec distributed job queue in Go
- 🧠 Work across systems programming, distributed systems, AI pipelines, and full-stack SaaS
- 🐧 Linux enthusiast who reads kernel docs for fun
→ Redis-backed job queue built from scratch in Go
- Priority scheduling via lexicographic score encoding on Sorted Sets
- Atomic Lua-scripted dequeue — eliminates race conditions under concurrent load
- Retry engine with exponential backoff + jitter, Dead Letter Queue, job replay
- Prometheus metrics: queue depth, job duration histograms, worker utilization
- 18,800 req/sec with 100K jobs, 100 concurrent workers — p99 latency 33ms, zero failed requests
🔗 https://github.com/Aryan9inja/gotaskq
→ Event-driven HTTP server built using POSIX sockets and epoll, zero external libraries
- Non-blocking I/O with request state machine
- Keep-alive, zero-copy sendfile(), request pipelining
- Full GET/POST parsing, headers, static file serving, EAGAIN/EWOULDBLOCK handling
- 14,000+ req/sec, sub-millisecond latency — 38% throughput improvement over fork model
🔗 https://github.com/Aryan9inja/raw-http-c
→ Farmers call one number, speak Hindi or English, get crop/pest/mandi guidance in under 2 seconds
- Twilio Conversation Relay for real-time STT/TTS
- Hybrid BM25 + vector retrieval over 1,000+ knowledge entries — sub-millisecond query latency
- Swappable LLM client (Groq/OpenAI), stateless WebSocket layer
🔗 https://github.com/Aryan9inja/Krishi-Setu | Demo
→ RAG-based platform that generates and reviews legal documents
- Classifies every contract clause into Low/Medium/High risk — cuts manual review time by 60%+
- MongoDB Atlas Vector Search for semantic retrieval
- Live in production with PDF export
🔗 https://github.com/Aryan9inja/LexGenAI | Live
Full-Stack Developer Intern @ SvaraAI (Oct 2025 – Mar 2026) London-registered B2B SaaS startup — Remote
- NestJS/Node.js microservices handling 1,000+ API requests/day across 5+ business clients
- Production email pipeline via Azure Email Services — 500+ emails/day, SPF/DKIM/DMARC configured
- AWS EC2 + NGINX + MySQL infrastructure, CI/CD with GitHub Actions
Languages: C • C++ • Go • JavaScript • TypeScript
Backend: Node.js • NestJS • Express • REST APIs • WebSockets • Microservices
Frontend: React • Next.js • TailwindCSS
Databases: MySQL • MongoDB • Redis • Vector Search
AI / Data: RAG Systems • BM25 Retrieval • LLM Integration (OpenAI, LLaMA) • Semantic Search • OCR
Infrastructure: AWS EC2 • NGINX • Linux • Docker • Prometheus • Azure Email Services • GitHub Actions
- 🚀 Fascinated by Space & Astronomy
- 🧩 Enjoy reverse engineering and understanding software internals
- ☕ Can debug almost anything with enough coffee
⭐ I build systems to understand how software really works — from the kernel up.