Hi HN!
Really grateful for the feedback and users over the past couple of months, the Show HN community has been so helpful! We've recently improved some core pain points affecting our Ubik users. What's New? 1. Improved PDF ingestion: wait times have been reduced, varying times based on size & # of PDFs uploaded (DOCX soon, wait for files to turn green in context bar). 2. Simplified layout: Panel option added (top right of page). 3. Added models: 200+ frontier/experimental models. 4. Compare Canvas: See text documents (Canvas) side by side, apply changes where you see fit. 5. Improved system prompt/tool calling: Agents are more aware of all tool purposes. 6. Abstract page: not all papers are open-access - new abstract page has core info and routes to DOI. What is Ubik? Ubik is an AI Research Environment (AiRE). With context-aware agents, academic database access (ArXiv/Semantic Scholar), and consistent human approval, Ubik improves agentic capabilities, minimizes hallucination, and makes LLM-assisted writing usable in high-level research and citation-based work. How does Ubik do this? Agent Orchestration We build our agents optimized around their environments, not a specific model, making it easy to run them in the cloud, on devices, and eventually with local models. Dynamic Context Engine This helps the AI and the human understand the larger project at hand, as well as any research criterion (citation styling, topic area focus, finding new sources, context of a project) 1. NOT just embeddings or semantic search! 2. Custom document parsing and enhanced OCR for in-text citations and transparent AI output 3. Agentic analysis to extract, understand, and markup documents beyond single questions 4. Like Cursor, Ubik knows your workspace. For improved accuracy, you can reference all files and agent-made items (notes/canvas) while prompting using the @ symbol. Why did we build Ubik? Because generative tools help experts and hurt beginners, quick solutions to complex problems without friction/interactivity between the initial prompt and the desired output don’t build cognitive skills in beginners that experts have mastered without AI (vibe code, vibe physics, vibe xyz = AI slop). Unlike beginners, experts amplify their human intelligence because they can correct AI output and understand when the model/agent is unhelpful or hallucinating; this is intellectual agility, a critical skill for effective AI use. https://stackoverflow.blog/2025/08/07/a-new-worst-coder-has-... https://simonwillison.net/2025/Oct/7/vibe-engineering/ https://bigthink.com/starts-with-a-bang/vibe-physics-ai-slop... The Cognitive Cost: MIT is conducting an ongoing neurological study on ChatGPT users' brain activity while using the AI chatbot to complete tasks. MIT's early findings are worrying. LLM-assisted writers could not complete writing tasks effectively or efficiently. What does this mean? Critical thinking skills and intellectual agility crumble when we treat AI as an oracle. https://www.media.mit.edu/projects/your-brain-on-chatgpt/ove... https://bigthink.com/business/ai-will-never-be-a-shortcut-to... Ubik v1 Today: https://app.ubik.studio is live (v1). Our current 260 users (researchers, scientists, and students) who use Ubik for their citation-based work. Before we release the full Ubik App, the platform will be free! Sign up for our PDF analysis tools, and please reach out with any feedback. Ubik App v2 (electron app) Coming Soon: Reduces all wait times on PDF ingestion, unifies our agents, and is built ready for local, on-machine models. We are working on our custom eval suite for knowledge work/evidence attribution (reach out to learn more). If you are interested in a demo of our full app, would like to enter a beta/testing group for the app, or want to understand more about Ubik, please email me: ieuan@ubik.studio |