Hi, I'm Rahul
RSS FeedI'm a machine learning engineer at ACV Auctions in Chennai, India. I build production ML systems end to end: data pipelines, training, evaluation and serving.
Before that I worked on ML inference for TI TDA4 edge chips at MulticoreWare, and on computer vision at TÜV Rheinland, where I cut model inference from 272 ms to 41 ms. My master thesis at TUM and PreciTaste was CLIPtheCenter, few-shot object detection with CLIP.
Co-author of PhoCaL (CVPR 2022), a multi-modal dataset for category-level object pose estimation. Here I write about what I learn and keep a record of projects. More about me.
Selected Projects
-
CLIPtheCenter
Master thesis: few-shot object detection by extending CenterNet with CLIP and DINO class codes.
-
LLM Explorer
Retrieval-augmented generation with LangChain, synthetic data from TinyLlama, and Hugging Face inference experiments.
-
PriorDepth
Joint monocular depth and camera-motion estimation from video, combining KeyPointNet and MonoDepth2.
-
Product Photography with Stable Diffusion and LoRA
Fine-tuning Stable Diffusion 1.5 with LoRA for product photos, filtering products with SAM and CLIP, and refining them with ControlNet inpainting.