Back to Tech Stack

RunPod

Cloud GPU platform for AI and ML workloads with instant access to high-performance GPUs for model training and inference.

RunPod

Cloud computing platform specialized in GPU infrastructure for AI/ML workloads. Provides instant access to high-performance GPUs at competitive prices.

Key Features

  • GPU Selection: Wide range of GPU options (A100, H100, RTX series)
  • Instant Deployment: Quick spin-up of GPU instances
  • Flexible Pricing: Per-hour or long-term rentals
  • Pre-configured Templates: Ready-to-use ML environments
  • Serverless GPU: On-demand GPU execution

My Experience

Used RunPod for renting GPUs during the LLM fine-tuning process for Chandra X-ray Observatory data analysis, enabling cost-effective access to powerful compute resources.

Projects Using RunPod

Fine-tuning LLMs for Chandra X-ray Observatory Data

Astromind • December 2024 - Present

How I used it: We used runpod for renting GPUs for finetuning process

Contrastive Learning: Aligning Chandra Event Data with Research Papers

Astromind (in collaboration with CfA Harvard) • December 2024 - Present

How I used it: For renting GPUs

PLM: Physical Language Model for Chandra X-ray Source Analysis

Astromind • 2024 - 2025

How I used it: GPU infrastructure for model fine-tuning and inference