Posts by Teemu Karkkainen
VLM Fine-Tuning for Robotics on AMD Enterprise AI Suite
- 28 November 2025
Vision-language models (VLMs) power applications from image captioning to robotics instruction following, but full model fine-tuning is resource-intensive and slow. Low-Rank Adaptation (LoRA) offers a faster, more efficient alternative by training only a small set of injected parameters while keeping the base model frozen.
AMD Inference Microservice (AIM): Production Ready Inference on AMD Instinct™ GPUs
- 17 November 2025
As generative AI models continue to expand in scale, context length, and operational complexity, enterprises face a harder challenge: how to deploy and operate inference reliably, efficiently, and at production scale. Running LLMs or multimodal models on real workloads requires more than high-performance GPUs. It requires reproducible deployments, predictable performance, seamless orchestration, and an operational framework that teams can trust.