AI Tools for Scientists
A curated guide to tools that are actually worth your time
A curated list of AI-powered tools to assist with literature review, writing, data analysis, and teaching.
AI Philosophy: Our mission is to understand life and champion open science by sharing that knowledge with the world. We use AI to remove barriers to that mission, using it to code faster, write clearer, and think bigger. Yet, these powerful tools demand equally rigorous oversight. We use AI to expand our minds, not replace them, demanding 100% verification to ensure our work is transparent, reproducible, and stands on a bedrock of accuracy.
Frontier Models (The Engines)
These are not simple chatbots; they are reasoning engines. With the right prompting strategy (giving context, persona, and constraints), they can draft entire manuscripts, synthesize vast literature, and code complex pipelines. If you use them lazily, you get garbage. If you drive them like a Ferrari, you win.
-
ChatGPT (OpenAI)
The original heavy hitter. The "o1" model is currently the king of complex reasoning. If you are using the free version (GPT-3.5 or 4o-mini) for science, you are wasting your time. Pay for the real tools.
-
Claude (Anthropic)
Currently the best prose writer of the bunch. It sounds less "robotic" than the others and handles massive text inputs (like whole books) beautifully. If you need to draft a paper, start here.
-
Gemini (Google)
Fully integrated into the Google ecosystem and has a massive context window. Can be brilliant at multimodal tasks, but can also be confidently wrong. Don't let it do your thinking for you - check its work. Currently this is my go-to for many tasks.
The power of these tools is highly dependent on your prompting. See my guide to prompting - especially the "mega-prompts" section, which is where the real capability is unlocked.
Prompt Engineering (Tools to Build Tools)
If you are building complex workflows (mega-prompts), you cannot just type into a chat window. You need environments that let you test, iterate, and save your "source code."
-
Anthropic Workbench
The best place to build prompts. It has a "Generate a Prompt" feature where you describe a task, and it writes a structured mega-prompt for you. Essential for system prompt testing.
-
OpenAI Playground
Stop using the normal ChatGPT interface for complex tasks. The Playground lets you edit the "System" message separately from the "User" message, which is critical for getting the AI to stay in character.
-
Fabric (GitHub)
An open-source project by Daniel Miessler. A massive library of "Patterns" (mega-prompts) for analyzing life, science, and security. A goldmine for seeing how experts structure their requests.
-
PromptPerfect
Takes your rough prompt and rewrites it into a detailed, optimized version. Good for learning how to be more specific.
-
TypingMind
A much better UI for ChatGPT/Claude. Lets you create a "Prompt Library" where you can save your favorite mega-prompts and call them up with a slash command (e.g., /review).
Literature Search & Discovery
-
Google NotebookLM
AI research assistant that lets you upload your own PDFs and sources to answer questions and summarize content strictly from your material. Can generate podcast-style audio summaries of documents you provide.
-
Elicit
Automates literature reviews by finding relevant papers and summarizing key takeaways in a table format.
-
ResearchRabbit
Visualizes citation networks to help you discover new papers relevant to your seed collection.
-
Consensus
AI-powered search engine that extracts and aggregates answers to scientific questions directly from peer-reviewed literature.
-
Scite
Smart citation tool that classifies citations to show whether a paper supports, mentions, or contrasts with the cited claim.
-
Google Scholar AI
Gemini chat with access to Google Scholar. Great for finding papers that speak to a complex idea or question.
-
Google Scholar PDF Reader
Chrome extension that adds an AI sidebar to online PDFs to generate outlines and summaries while you read.
-
Connected Papers
Creates a graph of similar papers based on co-citation and bibliographic coupling.
-
Scholarcy
Breaks down complex papers into digestible flashcards, highlighting key findings and methodology.
Writing & Editing
-
Paperpal
AI writing assistant trained on academic manuscripts to check grammar and style against journal standards.
-
Writefull
Provides language feedback using a database of published scientific papers to ensure academic tone.
-
DeepL
Highly accurate AI translation tool useful for clarifying complex phrasing.
-
Grammarly
Standard tool for catching typos, tone issues, and basic grammatical errors.
Data Analysis & Coding
-
GitHub Copilot
AI pair programmer that integrates into code editors (RStudio/VS Code) to autocomplete code.
-
Julius AI
Data analysis assistant that lets you upload datasets and ask questions in plain English to generate plots.
-
RTutor
Generates R code and plots by asking questions in natural language; great for syntax checks.
Figures & Visualization
-
BioRender
Industry standard for creating professional biological diagrams and pathways.
-
Adobe Firefly
Generative tool integrated into Adobe apps to create vector graphics and illustrations from text.
-
Midjourney / DALL-E
Generative AI tools for conceptual cover art or presentation backgrounds.
Teaching & Presentations
-
Gamma
Generates polished slide decks and presentations from a simple text outline.
-
Curipod
Interactive presentation tool that uses AI to generate polls and questions to engage students.
-
Quizlet
Automatically generates flashcards and practice tests from lecture notes.