The scientific community is embracing a new generation of research tools and networking platforms that change how researchers organise, collaborate, and share knowledge. From AI-powered research automation to advanced submission management platforms, these tools are redefining the way science is communicated and connected. They not only streamline workflows but also support global collaboration, making research more efficient and more impactful than it has been.
This guide isn’t a generic list of fifty apps. It’s an opinionated, ready-to-use shortlist covering each stage of the research lifecycle — along with the reasoning for why each tool earns its place over the alternatives, so you can choose deliberately rather than by default.
1. Literature Survey and Paper Tracking
Most researchers still rely on Google Scholar alerts, which are generally noisy, and Twitter/X is no longer the scientific commons it once was. The tools below make literature discovery considerably more meaningful.
Research Rabbit
If you want a genuine upgrade to paper exploration rather than another trial, this is the one-stop option. Think of it as the “Spotify for papers”: you pick a few articles in your domain and it returns a visually interconnected map of related citations, surfacing work you would never have found by keyword.
This makes literature review exploratory and connective in a way that a flat keyword list simply is not. Where tools like Connected Papers allow only a single paper as input and sit behind a paywall, Research Rabbit evolves continuously with your search and builds an entire collection for free. It also integrates cleanly with Zotero for reference management.
Semantic Scholar
Semantic Scholar is another free tool that complements the above rather than competing with it. Where Research Rabbit takes you deep into a specific topic, Semantic Scholar gives you breadth when you’re entering an unfamiliar field, surfacing influential citations with quick summaries so you can orient yourself fast.
Axy
Axy goes beyond helping you find papers — it helps you understand an entire research field you’ve just entered.
You can use AI to organise your lab’s scientific literature (published papers, preprints, or unpublished work) into a knowledge graph, then explore relationships between concepts, mechanisms, pathways, proteins, diseases, and other entities using natural language. That gives you an edge over citation graphs and keyword search simultaneously: you can ask questions that span multiple papers and receive evidence-backed reasoning pathways rather than a single AI-generated answer.
Conflicting findings are preserved rather than smoothed over, which makes it easier to spot genuine gaps in the literature and areas that need further work. Every change is reviewed and approved by a human before it becomes part of the shared knowledge base, so the graph stays transparent, auditable, and trustworthy. If graph figures are new to you, our guide on how to read a knowledge graph is a good starting point.
Elicit and Consensus
If you’re a PhD student or postdoc who already has field expertise and simply wants to stay current with literature synthesis, these are the go-to tools. Both provide structured literature searches driven by natural-language questions such as “What is the evidence for X?“
2. Managing References
Reference managers are common and well established, yet many researchers still cite manually or with outdated tooling. This category has three serious players — Zotero, Mendeley, and Paperpile. If you haven’t committed to one, here’s how to choose.
Zotero
Free, open-source, and with unlimited local storage, Zotero integrates cleanly with browsers, Word, Google Docs, and most academic workflows. That combination makes it the sensible default.
Zotero also has a rich plugin ecosystem — ZotFile, Better BibTeX, the Zotero–Research Rabbit sync — which gives it the most flexibility of the three. Mendeley is still widely used, but it hasn’t developed much over the years.
Paperpile
If you live in the Google ecosystem, Paperpile is the better fit. It has a cleaner interface and frictionless Google Drive integration, but it comes with storage constraints, less customisability than Zotero, and a mandatory paid subscription.
3. Reading and Annotation
Academic reading starts with a pile of downloaded PDFs and the time to work through each one. Rather than a basic PDF viewer, these tools make the process considerably faster.
SciSpace and Scholarcy
Both provide AI-powered paper summaries from just a DOI or an uploaded PDF. Getting key findings alongside methods and limitations in a single click makes it far quicker to triage which papers deserve a full read. SciSpace goes further with an “ask the paper” interface that is genuinely useful for unpacking dense methods sections.
Built-in Annotation and Hypothesis
Annotation can be handled directly inside your reference manager — both Zotero and Paperpile let you highlight, comment, and take notes on papers already in your library. Use Hypothesis if you want to annotate web pages as well as PDFs and share selected comments with your team, since it keeps public and private annotation layers separate.
4. Writing and Collaboration
Writing is where the work actually comes together, turning a sea of experiments into a coherent argument across papers, grants, and co-author revisions.
Overleaf
For LaTeX users who want to edit together in a browser, Overleaf is the gold standard. It’s the first choice for PhDs across physics, mathematics, computer science, and the wider STEM fields — even the free tier offers real-time co-authoring, version history, and direct journal submission integrations. Google Docs remains a reasonable choice for non-LaTeX researchers who value ease of use.
Notion and Obsidian
Both are used for personal journalling, notes, and idea management, and the choice comes down to preference: local, link-rich notes (Obsidian) versus a cloud-connected database (Notion). Obsidian helps a research student build a personal knowledge graph over time, while Notion is stronger at connecting ideas across papers and projects through bidirectional links.
5. Journal Clubs and Lab Meetings
Journal clubs remain one of the best ways for research groups to stay current, but coordinating papers, discussions, and notes gets disorganised quickly. Extracting durable ideas from a journal club is frequently the goal and rarely the outcome.
Zotero Groups
If your lab already uses Zotero, Zotero Groups is among the easiest ways to run a journal club. Members maintain a shared library, organise readings into collections, and work from the same references.
Create a shared “Journal Club” collection where members add papers through the month before one is selected for discussion. It’s free, collaborative, and integrates with the Zotero libraries people already have.
Notion
Many labs use Notion as a central workspace for journal club schedules, discussion notes, reading lists, and action items. It’s highly customisable and easy to collaborate on across a research group.
The real payoff is a searchable archive of presentations, summaries, and follow-up experiments, so insights aren’t lost the moment a meeting ends.
6. Conferences and Networking
Conferences are where researchers come together to collaborate, keep up with developments in their field, and generate new ideas.
Whova
Whova has become one of the most widely adopted conference platforms in academic meetings, helping attendees navigate schedules and take part in event discussions.
It centralises schedules, announcements, messaging, and networking in a single app. Attendees can build personalised agendas, receive session updates, and connect with other researchers before and during the event.
Ex Ordo
Ex Ordo provides a central hub where researchers submit work, reviewers evaluate hundreds or thousands of papers, and organisers build schedules and structured programmes. It replaces email-based submissions and spreadsheets with a transparent, efficient system, and supports multi-track agendas — which makes it well suited to large conferences with diverse sessions.
For attendees, it also offers personalised schedules and clear visibility into accepted presentations. Its integration with networking tools such as Whova and Swapcard makes it a fairly complete solution, combining scholarly management with professional engagement.
7. Data and Code
No single tool serves every field here, but a few have become near-universal.
GitHub
GitHub is the non-negotiable backbone for computational research. Version control for code, scripts, and even datasets (via Git LFS) is table stakes for reproducibility, and it lets you track every change to an analysis pipeline.
Jupyter Notebooks
The standard for interactive computing in Python, R, and Julia — particularly useful for exploring data and sharing reproducible analyses alongside the explanation of what they show.
Weights & Biases and Neptune
These are the leading experiment trackers for machine learning research. Weights & Biases is strong on logging hyperparameters, metrics, and model artefacts across runs, while Neptune leans more towards collaboration and the model registry.
8. Science Communication and Visibility
Science is validated when it is seen and discussed. A published paper with no readers has no impact, which makes visibility part of the work rather than an afterthought.
Bluesky
Bluesky has emerged as an academic hub for science communication following the decline of Twitter/X. Researchers, journalists, and science communicators migrated there through 2024 and 2025, and the platform has developed meaningful discussion and network effects across a range of fields. It’s a lower-noise way to stay current in your area.
ORCID
ORCID gives every researcher a persistent identity across digital platforms. An ORCID iD links your grants, publications, and affiliations regardless of changes to your name or institution. It takes under five minutes to create, is a required step for most journal submissions, and is worth sharing at conferences to stay connected to other people’s work.
The Bigger Problem: Scientific Knowledge Is Fragmented
Researchers today have excellent tools for discovering papers, managing references, annotating PDFs, writing manuscripts, and analysing data. But there’s one significant gap.
The knowledge itself isn’t shared.
Each researcher builds a mental map of their field by reading hundreds of papers, connecting ideas, and identifying patterns. That map rarely exists outside their own notes. Collaborators can’t benefit from it, and it usually disappears when the researcher leaves the lab. This isn’t a productivity problem — it’s an infrastructure problem.
The next generation of research tools won’t just help researchers read papers faster. They’ll help research groups build shared, living representations of scientific knowledge that evolve alongside the literature. Knowledge graphs are a step in that direction, and AI is already reshaping how they’re built.
By combining AI-assisted extraction with human review, transparent evidence, and collaborative editing, platforms like Axy turn scattered publications into an auditable, continuously improving map of scientific understanding.
Call to Action
The tools above will make you faster at every individual stage of research. What none of them do on their own is leave behind a shared map that outlives the person who built it. Axy is built for that gap — folding your lab’s literature into a scientific knowledge graph that surfaces connections across papers, preserves conflicting evidence, and stays auditable because every change passes human review.
If you want to help build the shared, connected map of science rather than just use another tool, apply to join below.
