Saujan Bohara

Adaptive Inference A Metrics-Based Gateway for vLLM

Saujan Bohara

Inside My GPU Inference Lab 🧠 From Staging to Production on Kubernetes ☸️

Saujan Bohara

Adaptive Inference Router (FastAPI + Redis) Protect vLLM GPUs Under Load 🚦

Saujan Bohara

Why K8s DNS Fails Across Tailscale — Calico IPIP vs VXLAN Fix

Saujan Bohara

Building a Hybrid Kubernetes Cluster for GPU Workloads (Local + Vast.ai)

Saujan Bohara

Building a Model Router on Kubernetes (3 Models,1 Router)

Saujan Bohara

How to Monitor Key LLM Metrics (GPU + Grafana Dashboard)

Saujan Bohara

Building a vLLM Chat UI on K8s (FastAPI + Local LLM + Metrics)

Saujan Bohara

🚀 Practical vLLM Demo — Real GPU Performance Test

Saujan Bohara

Why vLLM Feels So Fast (3s vs 19.6s 93% vs 29% GPU)

Saujan Bohara

Kubernetes Pod vs Bare-Metal GPU VRAM & Utilization Compared

Saujan Bohara

Run AI Models Locally — Hugging Face + GPU Explained

Saujan Bohara

🔧 GPU Monitoring ServiceMonitor Deep Dive + Grafana Dashboard Setup

Saujan Bohara

🚀 Grafana + Prometheus on Kubernetes (GPU Node) — 1-Line Install Setup 🧠

Saujan Bohara

🧠 Setting Kubernetes cluster on a GPU node with NVIDIA Operator Vast.ai GPU Cluster Demo

Saujan Bohara

GPU Basics Provisioning + Debugging with NVIDIA-SMI

Saujan Bohara

What is a VRAM

Saujan Bohara

3 Types of GPUs Explained Consumer, Workstation & Data Center

Saujan Bohara

Which GPU Cloud is Best for AIML (SRE Perspective)

Saujan Bohara

Helpdesk to Cloud in 5 years

Saujan Bohara