Skip to content
.NET / distributed systems / Kafka / UDP-TCP sockets

.NET under real load,from APIs to sockets.

I own event-driven .NET services sustaining 1.5M requests an hour — REST contracts, MySQL schemas, RabbitMQ and Kafka messaging, and UDP/TCP sockets driving sortation hardware across three countries.

See production recordOpen resumeExplore the projects
Open to backend engineering rolesNoida, IndiaHybrid2+ years experienceB.Tech Computer Science · 2024UTC +05:30
PRODUCTION SIGNAL PATH / SORTATION ESTATELIVE

  1. INBOUND OPERATIONS

    Operations software

    Routing state enters the service boundary.

  2. .NET SERVICES

    1.5M+ API calls per hour

    .NET APIs coordinate each parcel decision.

  3. MESSAGING / DATA

    RabbitMQ · Kafka · MySQL

    Events move, durable state stays true.

  4. EQUIPMENT CONTROL

    UDP / TCP sockets

    Commands cross the machine boundary.

  5. SORTATION MACHINERY

    50k+ parcels / hour

    Decisions become physical outcomes.

SERVICE OWNERSHIP

Interface → persistence → on-call

DEPLOYMENT ESTATE

India · UK · Saudi Arabia

requests sustained per hour
1.5M+
parcels coordinated per hour
50k+
countries with live deployments
3
selected systems projects
4

OPERATING RECORD / PRODUCTION

Software on one end.Machinery on the other.

I own .NET services behind industrial sortation across India, the UK, and Saudi Arabia. The common thread is predictable behaviour when traffic, equipment, and operations all become real.

REAL-TIME DISTRIBUTED SYSTEMS

Software Engineer

Falcon Autotech

Sole service ownership across three countries

I own the .NET services standing between warehouse operations software and the machinery on the floor running in production at sites across India, the UK, and Saudi Arabia.

1.5M+ REQUESTS / HOUR
Build and operate .NET services sustaining 1.5M+ requests an hour across live production deployments.
SOLE SERVICE OWNERSHIP
Own multiple production services end to end: client-facing API design, MySQL data modelling, delivery, and on-call.
50K+ PARCELS / HOUR
Drive sortation hardware directly over UDP and TCP sockets, coordinating equipment that peaks above 50,000 parcels an hour.
EVENT-DRIVEN ARCHITECTURE
Decouple services with RabbitMQ across the estate, reaching for Kafka on the workloads whose throughput justified it.
WORKING STACK.NETC#MySQLRabbitMQKafkaUDP / TCP
INVENTORY OPERATIONSINTERNSHIP

Web Application Developer

Inxee Systems Pvt Ltd

End-to-end product delivery

Shipped a custom inventory system across the interface, API, and persistence layers.

25% FEWER DISCREPANCIES
Built inventory workflows that reduced stock discrepancies by 25%.
30% BETTER ORDER ACCURACY
Improved order accuracy by 30% through clearer stock and fulfillment state.
25% FASTER PAGE LOADS
Improved React interface performance, reducing page-load time by 25%.
FULL-STACK DELIVERY
Implemented Node.js and Express services with MongoDB persistence.
WORKING STACKReactNode.jsExpressMongoDB

Production establishes the constraints. The selected projects next expose the decisions behind traffic control, retrieval, caching, and multi-channel delivery.

Continue to projects

SYSTEMS PORTFOLIO / THREE LABS + ONE BUILD

Four systems.Decisions exposed.

Three controlled labs make backend decisions operable. One detailed build shows how a multi-channel notification service is designed from intake to delivery.

Every lab runs in this site’s own codebase — a deterministic TypeScript engine behind each one, covered by unit tests. The featured build opens into a non-interactive engineering case study.

EDGE / 01TRAFFIC CONTROLLAB
QUOTA ARMED
press sendlive quota
3requests left
CLIENTEDGELIMITER
algorithm
token bucket
refill
2 / sec
responses
200 / 429

What happens when request 101 arrives?

A playable rate limiter: a live quota you can watch, requests you fire one at a time, and the exact reason each one came back 200 or 429.

3

algorithms under test

  • A real-time gauge of how many requests you can still make, and when the next one unlocks
  • One click sends one request along a visible pipeline, decided the moment it reaches the limiter
  • Token bucket, fixed window, and exact sliding log, each explaining itself in its own numbers
Spend the quota
NEXTJSTYPESCRIPTSVGREQUESTANIMATIONFRAME
AI / 02RETRIEVAL SYSTEMSLAB
INDEX READY
AI / 023 passages

Query

How does grounding reduce hallucinations?

Retrieved contextScore
RAG grounding0.98
Source attribution0.93
Fallback policy0.81

Why did this answer trust that passage?

A browser lab for inspecting how a query becomes ranked context and then an answer — on your own documents.

384

embedding dimensions

  • Bring your own PDF, DOCX, TXT or pasted text — chunked, embedded and searched without ever leaving the browser
  • Real 384-dimension embeddings from all-MiniLM-L6-v2 running on WebGPU/WASM via transformers.js in a worker
  • Optional local answer generation with SmolLM2-360M on WebGPU, with a copyable grounded prompt as the fallback
Run a retrieval trace
NEXTJSTYPESCRIPTTRANSFORMERS.JSWEBGPU
DATA / 03CACHING SYSTEMSLAB
CACHE WARM
DATA / 03cache-aside
user:1hit
18s
product:2hit
09s
order:1evict
LRU
cart:4miss
DB
ClientCacheDatabase

What changes when a cache hit becomes a miss?

An interactive cache-aside simulator with visible hits, misses, expiry, and eviction.

5

cache events exposed

  • Animated request flows between client, cache node, and database
  • Cache-aside strategy with hits, misses, invalidation, TTL countdowns, and LRU eviction
  • Live metrics: hit rate, DB reads/writes, evictions, and modelled latency
Explore the cache
NEXTJSTYPESCRIPTTAILWIND CSSSVG
MESSAGING / 04NOTIFICATION INFRASTRUCTUREBUILD
CASE STUDY READY
MESSAGING / 04case study

One publish contract

IntakeQueueWorkers
urgentpriority
standardqueued
bulkthrottled

Channel adapters

EMAILSMSPUSHWEBHOOKSLACKTEAMSWHATSAPP

How do seven channels share one delivery contract?

A high-throughput, queue-based notification service designed to dispatch more than one million notifications per day across seven delivery channels.

1M+

notifications designed per day

  • One publish contract fans out to email, SMS, push, webhook, Slack, Teams, and WhatsApp adapters
  • Priority queues, scheduling, and rate limits keep urgent traffic moving without overwhelming downstream providers
  • Exponential-backoff retries, dead-letter isolation, templates, and delivery analytics make failures inspectable
Read the case study
QUEUE-BASED DELIVERYPRIORITY QUEUESEXPONENTIAL BACKOFFDLQ

The labs expose system behaviour. The case study records the architecture, delivery path, and reliability decisions behind the build.

GitHub profile

CAPABILITY MATRIX

The whole path,not a logo wall.

Technologies matter in context: the traffic they carry, the state they protect, and the operators who depend on the result.

Services and APIs

Production services, clear contracts, and predictable behaviour under real traffic.

  • .NET
  • C#
  • Node.js
  • REST
  • Next.js

Messaging and data

Durable state and event flows selected for the workload rather than the trend cycle.

  • RabbitMQ
  • Kafka
  • MySQL
  • MongoDB
  • SQL

Real-time integration

Software that crosses the network boundary and coordinates physical equipment.

  • UDP
  • TCP
  • Equipment control
  • Observability

Runtime and delivery

Packaging services and running them the same way off the local machine.

  • Docker
  • Linux

Interfaces and delivery

Enough frontend depth to make backend state inspectable, useful, and testable.

  • TypeScript
  • React
  • Tailwind CSS
  • SVG
  • Vitest

FIELD NOTE / THE OPERATOR

I like systems thatexplain themselves.

I build software where the result is visible in the real world: services that coordinate sortation equipment, APIs that stay predictable under load, and project case studies that expose the decisions inside a backend system. I enjoy owning the whole path from interface and protocol to persistence and production.

NEXT COMMISSION / DIRECT LINE

Need an engineer whoowns the whole path?

I’m looking for backend and software engineering roles where reliability, systems thinking, and clear ownership matter.

Open to backend engineering roles[email protected]

Developed by Abhishek Suman with