Hello
Here's who I am
and what I do
Versatile AI Engineer with a background in developing deep learning models and deploying AI in products. I authored two open-source libraries. EmbedAnything has over 300,000 downloads and 1300 GitHub stars. I hold an MSc. in Artificial Intelligence from Eindhoven University of Technology (Cum Laude) and have six years of prior industry experience spanning machine learning, software development, and engineering.
Recent Blogs
Optimize VLM Tokens with EmbedAnything and ColPali
ColPali is a late-interaction vision rag model that enables text search within images. It can locate specific pages in PDFs containing relevant text, even when the text exists as part of an image. This powerful capability forms the foundation of Vision RAG systems.
Build ReAct Agents using SLMs from Scratch
Build a function-calling agent using a custom Unsloth model with a LoRA adapter, switching between reasoning and function execution, following ReAct-style reasoning.
Using YOLO with CLIP to improve Retrieval
In this article we explore how object detection can be used with embeddings to make retrieval more better.
Featured Projects
EmbedAnything
A Rust-based library designed to optimize embedding pipelines, making them faster and more reliable.
- Local, ONNX, and cloud embedding model support
- Multimodal processing (text, images, audio)
- Vector streaming for memory-efficient indexing
- 91.2% Rust, 8.5% Python codebase
Reflect
Memory for AI agents that stores, ranks, and retrieves reflections so agents learn from outcomes and improve with every execution cycle.
- Long-term memory for AI agents
- Continuous improvement through outcome learning
- Works across any operating system
Pulsar
A Cursor-like web IDE for knowledge work. Write, research, and draft alongside AI with full visibility into every file change and tool call.
- Monaco editor with markdown preview and PDF viewer
- Streaming agent chat with diffs and thread history
- Agent Skills standard and multi-project workspaces
- Slack and Telegram integrations
Ajna DrawSense
AI-powered engineering drawing analysis. Upload technical drawings and automatically extract dimensions, tolerances, and machine fit checks.
- LangGraph + Gemini multi-stage analysis pipeline
- YOLO detection for dimension annotation regions
- Machine capability matching and tolerance validation
- React frontend + FastAPI backend on AWS Lambda
Contributor to Candle
Contributing to Candle, a minimalist ML framework for Rust with 16k+ stars. Added three key models to enhance the framework's capabilities.
- ColPali: Vision RAG Model
- Splade Model: Sparse Bert for BM25
- Reranker Model: Cross-encoder Architecture