{"id":"f7e614ed-5d55-4078-a807-6517548187dd","entityType":"agent","slug":"crewai-nabankur14-ai-research-assistant-crewai","name":"ai-research-assistant-crewai","canonicalUrl":"https://www.xpersona.co/agent/crewai-nabankur14-ai-research-assistant-crewai","canonicalPath":"/agent/crewai-nabankur14-ai-research-assistant-crewai","generatedAt":"2026-10-10T07:53:05.377Z","source":"GITHUB_REPOS","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-09T13:50:59.780Z","emptyReason":null},"description":"Multi-agent AI system that reads academic research papers and generates structured, publication-ready summary reports using CrewAI and Google Gemini. š§ ResearchMind AI Multi-Agent AI Research Assistant **Automated academic paper analysis, summarization, and quality review ā powered by CrewAI and Google Gemini** $1 $1 $1 $1
![\"ResearchMind](\"assets/AI_Research_Assistant_Thumbnail.png\")
\n\n# š§ ResearchMind AI\n### Multi-Agent AI Research Assistant\n\n**Automated academic paper analysis, summarization, and quality review ā powered by CrewAI and Google Gemini**\n\n[](https://www.python.org/)\n[](https://www.crewai.com/)\n[](https://ai.google.dev/)\n[](LICENSE)\n\n
\n
\n\n
pypdf]\n B --> C[š Research Analyst
Extracts objectives, methodology,
results & limitations]\n C --> D[āļø Academic Writer
Converts analysis into a
student-friendly report]\n D --> E[š§ Peer Reviewer
Audits accuracy, clarity
& readability]\n E --> F[š Final Structured
Research Report]\n\n style A fill:#1a1a2e,stroke:#4285F4,color:#fff\n style F fill:#1a1a2e,stroke:#34A853,color:#fff\n style C fill:#16213e,stroke:#6C5CE7,color:#fff\n style D fill:#16213e,stroke:#6C5CE7,color:#fff\n style E fill:#16213e,stroke:#6C5CE7,color:#fff\n```\n\n**Pipeline stages:**\n\n1. **Ingestion** ā the target PDF is loaded (from Google Drive in the reference implementation) and text is extracted from its pages using `pypdf`\n2. **Analysis** ā the Research Analyst agent extracts the paper's title, objective, problem statement, methodology, dataset, models, results, strengths, limitations, and future work\n3. **Synthesis** ā the Academic Writer agent transforms that technical analysis into a clear, well-structured educational report\n4. **Review** ā the Peer Reviewer agent checks the draft for technical accuracy, missing information, logical flow, grammar, and readability, rewriting sections as needed\n5. **Output** ā the final polished report is returned as plain text, ready to save, publish, or share\n\nThis is orchestrated using CrewAI's `Process.sequential` execution model, where a single `Crew` manages all three agents and tasks end-to-end.\n\n## Meet the Agents\n\n| Agent | Role | Responsibility |\n|---|---|---|\n| **Research Analyst** | Extracts technical substance | Reads the paper and accurately pulls out its core technical information |\n| **Academic Writer** | Simplifies without losing meaning | Converts dense technical findings into clear, educational content |\n| **Peer Reviewer** | Quality gatekeeper | Improves clarity, completeness, and factual consistency before final delivery |\n\n## Tech Stack\n\n| Category | Technology |\n|---|---|\n| **Multi-Agent Framework** | [CrewAI](https://www.crewai.com/) |\n| **LLM Interface** | [LiteLLM](https://www.litellm.ai/) |\n| **LLM Provider** | Google Gemini |\n| **PDF Processing** | pypdf |\n| **Language** | Python 3 |\n| **Development Environment** | Google Colab |\n| **Storage** | Google Drive |\n\n## Getting Started\n\n### Prerequisites\n\n- Python 3.10+\n- A [Google Gemini API key](https://ai.google.dev/)\n\n### Installation\n\n```bash\n# Clone the repository\ngit clone https://github.com/nabankur14/ai-research-assistant-crewai.git\ncd ai-research-assistant-crewai\n\n# Install dependencies\npip install -q -U crewai litellm pypdf\n```\n\n### Configuration\n\nSet your Gemini API key as an environment variable (or enter it securely at runtime ā the reference notebook uses `getpass` so the key is never hardcoded or persisted):\n\n```bash\nexport GEMINI_API_KEY=\"your-api-key-here\"\n```\n\n## Usage\n\n1. Place your target research paper (PDF) in your working directory or connected Google Drive\n2. Update the `pdf_path` variable to point to your file\n3. Run the notebook (or script) end-to-end\n4. The three agents will execute sequentially, printing progress as they work\n5. The final structured report is available via `result.raw`\n\n```python\nresult = await crew.kickoff_async()\nprint(result.raw)\n```\n\n## Sample Output\n\nTested end-to-end on the seminal paper *\"Attention Is All You Need\"* (Vaswani et al.), the pipeline produced a complete academic review, opening with:\n\n> *\"Demystifying the Transformer: An Academic Review of 'Attention Is All You Need'\"* ā covering the Transformer's departure from recurrence and convolution in favor of self-attention, its parallelization advantages, and its lasting impact on modern deep learning, followed by a full breakdown of introduction, methodology, results, and conclusions.\n\n## Repo Structure\n\n```\nai-research-assistant-crewai/\nā\nāāā notebook/ \nā āāā AI_Research_Assistant_(ResearchMind AI).ipynb # Main notebook: agents, tasks, crew, execution\nāāā README.md # Project documentation\nāāā LICENSE # License file\n```\n\n## Skills Demonstrated\n\n### Technical Skills\n       \n\n### Soft Skills\n   \n\n## Future Enhancements\n\n- [ ] Add OCR support for scanned/image-based PDFs\n- [ ] Remove the 12-page extraction limit for full-length paper support\n- [ ] Build a Streamlit front-end for non-technical users\n- [ ] Support multiple LLM providers (OpenAI, Anthropic, local models)\n- [ ] Export final reports directly to PDF/DOCX\n- [ ] Add a citation-extraction agent for reference validation\n\n## Author\n\n**Nabankur Ray**\nData Scientist & ML Engineer\n\n- š Portfolio: [analyticswithnaban.com](https://analyticswithnaban.com)\n- š» GitHub: [@nabankur14](https://github.com/nabankur14)\n- š LinkedIn: [@nabankur-ray](https://www.linkedin.com/in/nabankur-ray-876582181/)\n\n## License\n\nThis project is licensed under the MIT License ā see the [LICENSE](LICENSE) file for details.\n\n---\n\n\n\nā If you found this project interesting, consider giving it a star!\n\n
\n","readmeExcerpt":" š§ ResearchMind AI Multi-Agent AI Research Assistant **Automated academic paper analysis, summarization, and quality review ā powered by CrewAI and Google Gemini** $1 $1 $1 $1
pypdf]\n B --> C[š Research Analyst
Extracts objectives, methodology,
results & limitations]\n C --> D[āļø Academic Writer
Converts analysis into a
student-friendly report]\n D --> E[š§ Peer Reviewer
Audits accuracy, clarity
& readability]\n E --> F[š Final Structured
Research Report]\n\n style A fill:#1a1a2e,stroke:#4285F4,color:#fff\n style F fill:#1a1a2e,stroke:#34A853,color:#fff\n style C fill:#16213e,stroke:#6C5CE7,color:#fff\n style D fill:#16213e,stroke:#6C5CE7,color:#fff\n style E fill:#16213e,stroke:#6C5CE7,color:#fff"},{"language":"bash","snippet":"# Clone the repository\ngit clone https://github.com/nabankur14/ai-research-assistant-crewai.git\ncd ai-research-assistant-crewai\n\n# Install dependencies\npip install -q -U crewai litellm pypdf"},{"language":"bash","snippet":"export GEMINI_API_KEY=\"your-api-key-here\""},{"language":"python","snippet":"result = await crew.kickoff_async()\nprint(result.raw)"},{"language":"text","snippet":"ai-research-assistant-crewai/\nā\nāāā notebook/ \nā āāā AI_Research_Assistant_(ResearchMind AI).ipynb # Main notebook: agents, tasks, crew, execution\nāāā README.md # Project documentation\nāāā LICENSE # License file"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[],"languages":["python"],"docsSourceLabel":"GITHUB REPOS","editorialOverview":"Multi-agent AI system that reads academic research papers and generates structured, publication-ready summary reports using CrewAI and Google Gemini. š§ ResearchMind AI Multi-Agent AI Research Assistant **Automated academic paper analysis, summarization, and quality review ā powered by CrewAI and Google Gemini** $1 $1 $1 $1