Rudra Garg.

Backend systems.
Infrastructure. Applied AI.

I build services in Python and Go, from notification workflows and multiplayer backends to local AI and edge computing research.

IIIT Guwahati ’26
Former Backend Developer Intern at GRIG

Preparing the city…

PROFESSIONAL EXPERIENCE

Backend engineering at GRIG

Back End Developer Intern

June 2024 - August 2025 · Remote

About & Experience
  • Notification workflows: built alert.ventory with FastAPI and Temporal, including priority queues, retries, and human approval.
  • Inventory APIs: developed request.ventory with Firestore and Supabase, tenant-aware data handling, and GST and geographic validation.
  • Infrastructure and access: automated Kubernetes setup with Helm, Ingress, and cert-manager; implemented RBAC, OAuth2, and JWT sessions.
Read the notification workflow case study

Featured engineering projects

TerraQuest

Go · PostgreSQL · WebSockets · Vue 3

A multiplayer geography game with synchronized lobbies and server-side distance scoring. A Go worker pool validated 81,000+ Street View locations for its PostgreSQL dataset.

Engineering focus: concurrent lobby state, data ingestion, and indexed location queries.

Portal Gambit

FastAPI · React · Firebase · WebRTC

A chess variant with custom portal movement, shared game state, and peer-to-peer voice chat. Its FastAPI backend handles authentication, game history, and ELO analytics.

Engineering focus: asynchronous persistence, transactional relationships, and client time synchronization.

Research & applied AI

Model evaluations and simulation studies, with architecture notes and technical reports.

LOKI

A local voice assistant combining embedding-based intent matching, a local LLM fallback, and CRF entity extraction.

Evaluation context

Reported weighted NER F1: 0.9977. This measures entity extraction, not end-to-end assistant accuracy.

Read study & report

MUCEDS

Hierarchical reinforcement learning for UAV fleet sizing and positioning, with LSTM-based predictive caching.

Evaluation context

Simulation results: cache hits increased from 18% to 78%, with 53% lower average task latency against the study baseline.

Read study & report

Fog Computing Testbed

A four-tier Docker testbed for IoT module placement, with Linux network emulation and Prometheus/Grafana monitoring.

Evaluation context

The testbed compares sibling-aware placement with Edge-Ward placement under emulated network conditions.

Read study & report
Technical Arsenal

Stack Overview

Languages

Python
C++
SQL
Go
JavaScript

Backend & ML

FastAPI
Flask
Temporal
SQLAlchemy
PyTorch
NumPy
React
Vue

Cloud & Databases

AWS
Google Cloud
PostgreSQL
Redis
Supabase
Firebase
MongoDB

Infrastructure

Docker
Kubernetes
Helm
GitHub Actions
Prometheus
Grafana
Linux
Git