The Reading Hub: Transform Required Readings Into Your Living Research Workspace
Discover how Katav's Reading Hub unifies PDFs, notes, citations, and AI assistance into a single academic workspace. From undergrad syllabi to PhD dissertations, learn how to organize thousands of readings and never lose track of a source again.
William Zimmermann
Academic life is built on top of a quiet, relentless pile of PDFs. A first-year undergraduate might face fifteen articles a week across four courses. A master's student adds dozens of methodology chapters as a literature review takes shape. A PhD candidate ends up curating hundreds — sometimes thousands — of papers across a five-year project, where the difference between a citation found and a citation lost can be a missed argument, a weak chapter, or a delayed defense.
For most students, that pile lives in three or four disconnected places at once: a folder on the desktop, a tab graveyard in the browser, a few highlighted PDFs in a reader app, scattered notes in Notion or a notebook, and a syllabus PDF that nobody opens after the second week. Reading something is easy. Knowing what you've read, what you owe, what you've quoted, and where you wrote that one perfect line of analysis — that's where the system breaks.
Katav's Reading Hub exists to close that gap. It's the place where the syllabus, the references, the PDFs, the highlights, the page numbers, the citations, and the notes all live together — and where AI helps you read faster without reading less carefully.
What the Reading Hub Is
The Reading Hub is a structured workspace for academic readings, organized into three layers:
Reading Lists — collections of references, typically one per course, per subject, per chapter of a thesis, or per qualifying exam topic. Each list shows total items, items read, last activity, and a progress bar that quietly nags you toward the finish line.
Reading Items — the references themselves, each with full metadata (title, authors, year, type, DOI/URL, publisher, journal, volume, issue, pages, ISSN, PMID, citation key, language, library catalog, call number, and more). An item can be an article, a book, a chapter, or anything else.
Sections and Notes — inside each reading item, the student can divide a long paper or book into sections (Introduction, Methods, Chapter 3, etc.), each with its own page range and its own rich-text notes.
Around this spine, the Hub layers in PDF viewing, citation extraction, AI suggestions, AI summaries, and AI-assisted definitions.
The Core Features
Building a list, the lazy way
Creating a reading list takes two fields: a name and (optionally) a description. The list can be linked to a Course and a Subject already registered in Katav, so the Reading Hub stops being a parallel universe and becomes part of the same knowledge graph as the student's classes and notes.
Adding references can be done in four different ways, picking whichever is least painful at the moment:
Manual entry, with autocompletion on authors and tags that the student has used before.
Citation search through Crossref and PubMed, which fetches the canonical metadata (title, authors, year, DOI, abstract, keywords) directly from the source — no typos, no missing commas, no fabricated DOIs.
Bulk import from BibTeX-like reference lists, so an entire syllabus or a Zotero export can be pasted in at once.
AI-suggested readings: describe the research topic in plain language ("the influence of Habermas's public sphere on contemporary platform governance debates"), and the Hub returns a batch of candidate articles with title, authors, year, abstract, a short relevance explanation, and a link to the source. Each suggestion can be added to the list with a click, and "Load more" keeps the pipeline flowing without repeating itself.
Status, priority, tags — the simple triage system
Every reading item carries three pieces of state that, together, make the difference between a static bibliography and an actionable plan:
Status: Unread, In progress, Read. Counters at the top of the list keep score.
Priority: Required, Important, Complementary, Read later, Not important. This is what tells the student, on a Sunday night, which three of the eleven assigned papers actually have to be read by Tuesday.
Tags: free-form labels that travel across lists and become the connective tissue of a literature review.
The detail table sorts on any of these columns, and filters by status, so the same list can be a checklist before class, a triage view the week before an exam, or a reference index when writing a paper.
The reading workspace itself
Open a reading item and the screen turns into a three-panel academic cockpit:
Left panel — Metadata and actions: status picker, priority, tags, full bibliographic fields, DOI link, attached files, and the AI summary panel.
Middle panel — Sections: an outline of the paper or book, with page ranges and completion checkboxes per section.
Right panel — Notes editor: a TipTap-powered rich-text editor with headings, lists, tables, images, code blocks, highlights in multiple colors, sub/superscript, alignment, YouTube embeds, and — crucially — PDF citation blocks.
Attach a PDF to the item (stored privately on Vercel Blob) and a fourth panel slides in: an in-app PDF viewer with annotation support, page navigation, and a calibration tool that maps the PDF's numbering to the book's printed page numbers — the small piece of friction that ruins half of all academic citations. From the viewer, the student can select a passage and send it directly into the notes as a properly formatted citation block that remembers the page, the file, and the exact text. Clicking that citation in the notes later jumps the viewer back to the source page.
AI that reads alongside, not for, the student
The Reading Hub uses AI where it saves time without taking over the actual thinking:
AI abstract summary for the attached PDF, broken into four blocks — problem, method, result, limitations — so a student can decide in thirty seconds whether to spend the next two hours reading the whole paper.
AI Table of Contents extraction, which proposes sections from the PDF that the student can select and turn into structured sections in the workspace with one click.
AI section summary that compresses the student's own notes for a section into a clean recap — useful right before class or right before a review draft.
AI definitions for selected terms in the notes: highlight a phrase, click Define, and the system uses the surrounding paragraphs and the paper's title as context to produce a definition that's actually relevant to the field, not a generic dictionary entry. The definition can then be inserted into the notes with one click.
AI suggestions for further reading, scoped to a specific topic and capable of excluding what's already in the list to keep recommendations fresh.
Across all of these, the Hub is explicit that AI output can be wrong and surfaces a disclaimer so students don't outsource judgment they shouldn't be outsourcing.
How a Student Actually Uses It
The undergraduate: keeping up with the syllabus
For an undergrad juggling four or five courses, the Reading Hub becomes the answer to the question every undergraduate asks on Wednesday: "wait, what was I supposed to read for Thursday?"
A practical week looks like this. On day one of the semester, the student creates a reading list per course, linked to the corresponding subject in Katav. The professor's syllabus is bulk-imported as references. Each item gets a priority — Required for the actual assigned readings, Complementary for the recommended ones — so the list visually tells the student where to start.
Before each class, the student opens the relevant required reading, attaches the PDF, runs the AI abstract summary to orient themselves, then reads. Highlights live in the PDF; notes live in the right panel, structured by section. Quotes worth keeping are sent into the notes as PDF citation blocks, which means that later, when writing the term paper, the student can click the citation and instantly verify the source — including the right printed page number, not the PDF's internal page.
The class itself happens, and the notes from the Reading Hub sit next to the lecture notes in the same Katav workspace. At the end of the week, the list shows 7 of 9 readings as Read, the progress bar reflects reality, and the student knows exactly what to catch up on over the weekend.
The master's student: building a literature review
For someone writing a dissertation, the Hub turns from a checklist into a research instrument. A typical setup creates one list per chapter or per major theme of the literature review — Theoretical framework, Methodology, Empirical studies in Brazil, Empirical studies abroad, and so on.
References come in through Crossref or PubMed search (clean metadata is non-negotiable when a defense committee will check the bibliography), through bulk BibTeX import from earlier Zotero work, and through AI suggestions, which are particularly useful when a supervisor has just said "have you looked at the literature on X?" and the student needs to map the territory fast.
Each important paper gets the full treatment: PDF attached, AI abstract summary generated to triage relevance, sections defined per the paper's own structure, notes taken in the editor, key passages captured as PDF citations, and tags applied liberally — gender, foucault, qualitative, brazil-2010s — so that, three months later, a tag-filtered view can pull every paper relevant to a single argument.
When the writing starts, the student opens each list, reviews the AI section summaries to refresh memory, and pulls citations directly. Because every PDF citation in the notes remembers its file and page, the final manuscript's footnotes get built faster and with fewer errors.
The PhD student: a five-year knowledge base
At the doctoral level, scale becomes the problem. A PhD candidate accumulates references across coursework, qualifying exams, fieldwork, the dissertation itself, conferences, peer review duties, and side papers. The Reading Hub is structured for this.
Reading lists become functional zones — Qualifying exam: Theory, Qualifying exam: Methods, Dissertation: Ch. 2, Conference paper — ABA 2026, Peer review: journal X, Future projects — and references are shared across lists with the move/copy feature without losing their notes or annotations. A paper read for the qualifying exam in year one can be copied into the dissertation list in year three with all of its sections, notes, highlights, and PDF citations intact.
Long books — monographs, ethnographies, handbooks — fit the Hub particularly well. The candidate creates one reading item per book, then divides it into sections by chapter with their own page ranges. The page calibration tool aligns the PDF's numbering with the book's printed pages, so that every citation extracted from chapter 7 ends up referring to the right printed page in the final thesis. Each chapter section can be marked complete, and the AI section summary becomes a study aid before the qualifying exam.
When the candidate writes, the workflow is the inverse of what it used to be: instead of searching their hard drive for a PDF, then opening it, then searching inside it for the half-remembered quote, they search inside their own notes — which already contain the quote, the citation block, the page, the file. Click the citation, jump to the page, verify the wording, paste into the manuscript. Time saved per citation: a few minutes. Time saved over a 250-page dissertation: weeks.
Why It's More Than a Reference Manager
A traditional reference manager — Zotero, Mendeley, EndNote — stores citations. A traditional PDF reader — Apple Preview, Adobe, Foxit — stores highlights. A traditional note app — Notion, Obsidian, OneNote — stores text. The student is the integration layer between the three, which is to say: the integration layer is unreliable, badly indexed, and usually held together by a folder called to_read_FINAL.
Katav's Reading Hub collapses those three tools into one workspace where the reference, the PDF, the highlights, the notes, the page numbers, and the AI assistance all share the same context — and where everything is connected to the rest of the student's academic life: the courses, the subjects, the lectures, the class notes, the future exam study guides. The reference isn't just a citation in a list; it's a node in a knowledge base that grows with the student across semesters and across degrees.
For an undergraduate, that means never again losing track of what's due Thursday. For a master's student, it means writing a literature review with citations that actually point to the right page. For a PhD candidate, it means leaving the program with a research archive that's still readable, searchable, and reusable five years after the defense.
Reading is the unglamorous foundation of every academic project. The Reading Hub is the part of Katav that makes the foundation hold.