Queriously
Research copilot and technical document reader.
Queriously is a local-first desktop application designed for deep engagement with technical literature. It combines a high-performance PDF reader with a sidecar AI service to help researchers, students, and engineers understand complex documents faster.
Technical
decisions, with
defensible
evidence.
Read papers alongside an AI that annotates in the margins, cites sources, and runs entirely on your machine.
What ships in Queriously.
Marginalia Engine
AI-generated notes, contradictions, and simplifications rendered directly in the margins of the PDF.
Research Sessions
Named investigations that synthesize findings across multiple papers.
Grounded QA
A RAG-based chat engine with source-cited responses and page-level accuracy.
Local-First Privacy
All embeddings and AI processing (via Ollama) can run entirely offline.
Equations & Plots
Extracts and interprets complex equations and data plots from papers.
Traditional PDF readers are passive, and existing AI chat-with-PDF tools are disconnected from the reading experience, leading to context-switching fatigue.
Queriously was built to be a participant in the reading process. We moved the AI from a sidebar chat into the Marginalia — whispering insights in the margins exactly where the user is reading. Local-first architecture protects researchers’ intellectual property.
Researchers synthesize information across dozens of papers faster, with grounded AI assistance that reduces the cognitive load of reading dense technical literature.
Tech stack.
- Shell
- Tauri 2 (Rust)
- Frontend
- React 18TypeScriptZustand
- AI Backend
- Python 3.11 Sidecar (FastAPI)LiteLLMChromaDB
- PDF Engine
- pdfjs-distPyMuPDF
Have an idea, product, or business problem worth solving?
Let’s talk about how Growvth can help you build it. Tell us about your goals and we’ll get back within one business day.