TL;DR
AI research is bottlenecked by literature review, dataset discovery, and experiment tracking. MCPs cut all three: arXiv + Perplexity for paper discovery, HuggingFace for datasets and models, Pinecone or Supabase for embedding your own paper library, Jupyter for experiment execution, and GitHub for code. One Claude session replaces the usual 6-tool workflow.
The MCP stack for academic and applied AI researchers
A research scientist, PhD student, or applied ML engineer reading 5–20 papers a week, running experiments, and publishing insights.
What hurts today
- 1Reading 10 papers a week takes 15+ hours — most of it is skimming to decide 'is this worth reading deeply?'
- 2Finding the right dataset on HuggingFace or Papers With Code is slow and discovery-limited
- 3Reproducing a paper's experiment requires juggling the paper PDF, the code repo, and your own environment
- 4Writing related work sections is tedious because you need to re-read your own notes every time
- 5Sharing research with non-specialists means translating the abstract into a plain-English summary by hand
Recommended MCPs (7)
Perplexity
View MCPLiterature review on steroids — 'summarise the last 6 months of work on sparse attention with citations'. Cuts lit review from days to hours.
Tavily
View MCPBreadth search across arXiv, blog posts, and release notes. Useful for 'is there already a paper doing X?' before you invest in a research direction.
Firecrawl
View MCPScrape full paper PDFs (arXiv, openreview, papers with code) into markdown. Extract tables, equations, and related work from any paper.
Supabase
View MCPYour personal paper library with pgvector. Claude embeds every paper you've read and answers 'have we seen this technique before?' against your own history.
GitHub
View MCPClone repos referenced in papers, read the implementation, open PRs on your own research code. Critical for reproducibility.
Notion
View MCPYour research journal. Claude writes paper summaries, experiment logs, and idea threads directly into the right Notion databases.
Postgres
View MCPExperiment tracking — hyperparameters, eval metrics, checkpoints — all in Postgres. Claude queries your runs and compares trajectories without opening MLflow.
A real workflow
You're writing a related work section on long-context transformers. Claude uses Tavily to find the 30 most-cited recent papers on the topic, Perplexity to summarise the top 10 with key contributions, Firecrawl to pull full PDFs, Supabase pgvector to check which of these you've already read (it knows from your personal embedding library), and Notion to write the draft related work section. You get a structured 2-page draft with 30 proper citations in under an hour, where it used to take 2 days.
Time ROI
Researchers save 12–18h/week on literature review, paper summarisation, and related work. Downstream: you read 2x more papers and ship papers 2x faster.
Recommended recipes for this role
Research Automation
Paste a research topic in Notion and an agent uses Perplexity to gather sources, summarize findings, and structure them.
Web Scraping to Database
Schedule a Firecrawl scrape of any website and store the structured results directly in a Supabase table for analysis.
Search Results Indexing
Run Tavily searches on scheduled topics and index the results in Supabase for trend analysis and content research.
Competitor Watch Automation
Run daily Perplexity searches on competitors and log product updates, pricing changes, and news to a Notion tracker.
Knowledge Graph from Code
Parse your GitHub repos and build a Neo4j knowledge graph of files, functions, imports, and authors for code intelligence.
Frequently asked questions
Is there an official arXiv MCP?
A community-maintained arXiv MCP exists with search, download, and metadata tools. It's thin — most heavy lifting is done by Perplexity (discovery) + Firecrawl (PDF extraction). arXiv's API is open, so any MCP works.
What about HuggingFace?
The HuggingFace MCP covers models, datasets, and spaces. Critical for ML engineers: 'find me a text classification dataset with at least 100K examples in Portuguese' → answer in seconds.
Can Claude run my Jupyter notebooks?
Yes — the Jupyter MCP launches and executes notebook cells. Combined with the GitHub MCP (clone repo) and HuggingFace MCP (download model), you can reproduce most papers end-to-end from a single Claude session.
Is Pinecone better than Supabase+pgvector for a personal paper library?
For <100K papers, no. Supabase+pgvector is free up to 500MB and fast enough. Pinecone's advantage kicks in at 1M+ embeddings where latency and sharding matter. Most individual researchers never hit that.
Can Claude write paper reviews for me?
It can draft reviews — but the final review must be yours. Most reviewers use Claude for: initial summarisation, identifying key contributions, checking citations for completeness, and drafting the 'strengths/weaknesses' sections. The judgment stays human.
Other use cases
MCPs for SaaS Founders
A technical founder (0–10 employees) building a B2B SaaS who ships code, handles billing, writes marketing, and answers support — all in the same day.
MCPs for Solopreneurs & Indie Hackers
An indie hacker with a Twitter audience, a newsletter, 1–3 shipped products, and zero employees. Ships daily, markets constantly, avoids meetings.
MCPs for AI Agent Developers
A developer building AI agents, chatbots, or autonomous workflows. Needs search, scraping, vector storage, and LLM orchestration — all as tools the agent can call.
Start with this MCP stack
Install the full stack in one command, or cherry-pick the MCPs you need.
Browse all MCPs