Past battle · 2025-06-04 UTC
Research Assistants Showdown — June 4, 2025
From the Research Assistants category. 29 marks placed across 10 fighters. Toyota Woven City took the crown.
Final standings
The line-up
The fighters
Profiles of every tool that competed in this battle, ranked by their final score.
Toyota Woven City
Toyota's prototype smart city for testing AI, autonomy, and connected mobility at human scale.
Toyota Woven City is a purpose-built living laboratory at the base of Mount Fuji in Japan, designed by Toyota's Woven by Toyota subsidiary. It serves as a real-world testing ground where residents, researchers, and partner companies can develop and validate AI systems, autonomous vehicles, robotics, smart infrastructure, and sustainable energy technologies in everyday use. The city is organized around dedicated lanes for autonomous vehicles, pedestrians, and personal mobility, with sensor networks and a central digital operating system tying services together. Toyota positions it as a platform for collaboration with external inventors and startups, accelerating the path from prototype to deployment in connected urban environments. Rather than a consumer product, Woven City is best understood as an R&D environment and partnership program for organizations working on mobility, AI, and smart city solutions.
Criteria breakdown
- Dedicated autonomous vehicle roadways
- Connected sensor and data infrastructure
- Hydrogen and sustainable energy systems
- Robotics and in-home AI testing
- Digital twin of the physical city
- Partner program for startups and researchers
Harvey
An AI platform tailored for legal professionals, offering tools to streamline legal workflows and enhance productivity.
Harvey is an AI platform designed for legal professionals to streamline workflows and enhance productivity. It offers various tools such as an assistant for document analysis and drafting, a vault for secure document storage and analysis, and agents for executing complex legal work. The platform also includes features like contract intelligence, command center for analytics, and shared spaces for collaboration. Harvey aims to help legal teams navigate complexity, focus on high-value work, and deliver strategic value. The platform is used by top law firms and in-house legal teams for tasks such as legal research, deal management, due diligence, and contract analysis.
Criteria breakdown
- Assistant for document analysis and drafting
- Vault for secure document storage and analysis
- Agents for executing complex legal work
- Contract Intelligence
- Command Center for analytics
- Shared Spaces for collaboration
Semantic Scholar
AI-powered academic search engine for navigating scientific literature
Semantic Scholar is a free, AI-driven research tool developed by the Allen Institute for AI that helps users discover and understand scientific papers. It indexes hundreds of millions of publications across disciplines, using machine learning to surface the most relevant and influential work for any given query. Beyond keyword search, the platform extracts key insights from papers, generates TLDR summaries, maps citation networks, and identifies highly cited or seminal works. Researchers, students, and academics use it to accelerate literature reviews, track scholarly impact, and stay current in their fields. With an open API, personalized research feeds, and integrations with reference managers, Semantic Scholar serves both casual readers and serious research workflows.
Criteria breakdown
- AI-generated TLDR paper summaries
- Semantic search across 200M+ papers
- Citation context and influential citation tracking
- Personalized research feeds and alerts
- Author profiles and h-index metrics
- Public API and dataset for research use
PodSized generates 5-minute AI-powered summaries of podcast episodes, along with audio versions and full transcripts. The tool aims to help users skip filler content and retain key insights from podcasts across various categories such as business, tech, health, and science. PodSized processes hundreds of episodes daily and covers over 474 podcasts, including popular shows like the Lex Fridman Podcast, Huberman Lab, and The Tim Ferriss Show. The summaries are generated using AI and can be accessed in four ways: text summary, audio summary, full transcript, and timestamped chapters. The platform offers a free trial with no credit card required, and users can access summaries on iOS, Android, and web devices. While PodSized excels at providing concise summaries and audio versions, its capabilities may be limited by the quality of the AI model and the complexity of the podcast content. Overall, PodSized seems to be a useful tool for individuals who want to quickly grasp the main ideas of a podcast episode without committing to listening to the full episode.
Criteria breakdown
- AI-generated 5-minute audio summaries
- Curated library of popular podcasts
- Quick previews of episode content
- Topic-based browsing
- Mobile-friendly listening experience
Ticker Pulse is an AI-powered sentiment tracker that analyzes the voices shaping the markets, specifically finance YouTubers, to highlight trending tickers and their associated sentiment. It extracts mentions of stocks and cryptocurrencies from dozens of videos and posts daily, applying sentiment scoring from Extremely Bearish to Extremely Bullish. This allows users to discover which tickers are gaining traction and where the conversation started, with direct links to the original content. The tool aims to address the issue of information overload in the market, where stocks are influenced by both earnings and buzz created by creators. By automating the discovery process, Ticker Pulse helps retail investors find relevant content faster, catch early trends, and avoid hype traps by tracking sharp sentiment shifts. Ticker Pulse is designed for retail investors, whether trading daily or building a long-term portfolio. It saves users hours by providing a clear view of which creators are discussing specific tickers, enabling them to validate their ideas against the real-world crowd mood. The tool's standout capability is its ability to quantify the psychology of the market, connecting users with the voices driving it. By doing so, it provides an edge in a world where information overload can be costly. The Ticker Pulse dashboard offers a centralized platform for users to explore the market's pulse, with links to the original videos or articles behind the trends. By leveraging AI-powered sentiment analysis, Ticker Pulse provides a unique insight into the market, helping users make informed investment decisions.
Criteria breakdown
- Daily automated scanning of finance YouTubers and posts
- Ticker extraction for every stock and crypto mentioned
- Sentiment scoring from Extremely Bearish to Extremely Bullish
- Direct links from each mention to the original creator/source
- Time-range views: 7, 30, 60, or 90 days
- Daily Report with mentions count and average sentiment
WhispriNote
Turn voice, audio, and YouTube links into structured study notes with flowcharts.
WhispriNote is a study-focused transcription tool that converts spoken input, uploaded audio files, and YouTube videos into organized, readable notes. Instead of producing raw transcripts, it formats the content into summaries, key points, and visual flowcharts designed for learning and review. The tool is aimed at students, researchers, and self-learners who want to capture lectures, podcasts, or online videos and quickly turn them into reference material. By combining transcription with structuring and diagram generation, it reduces the manual effort of cleaning up notes after a recording session.
Criteria breakdown
- Voice-to-notes transcription
- Audio file upload and processing
- YouTube link import
- Automatic flowchart generation
- Structured study-ready output
- Summarization of key concepts
CurateIt
An AI-powered platform that streamlines content curation, bookmarking, and productivity for professionals and enthusiasts.
CurateIt is an AI-powered platform designed to streamline content curation, bookmarking, and productivity for professionals and enthusiasts. As a digital tool, it provides a space for users to save, organize, and access relevant content from various sources, facilitating informed decision-making and enhanced work efficiency. By automating tasks and leveraging AI technology, CurateIt aims to simplify the process of staying updated and inspired, making it easy for users to prioritize their time and tasks effectively. The tool can be used across diverse industries and fields, with both individuals and teams benefiting from its capabilities. While its AI-driven features and intuitive interface are expected to improve user experience, the platform's effectiveness and scalability will depend on its ability to continuously adapt to evolving user needs and technological advancements.
Criteria breakdown
- AI-powered content curation and organization
- Bookmarking and saving features for relevant content
- Task automation for improved productivity
- Intuitive interface for user-friendly experience
- Scalable platform for diverse user groups
&AI
AI workspace for patent litigators to build claim charts, prior art, and trial documents.
&AI is a platform for patent litigators to craft trial-ready work product. It helps users build claim charts, prior art, and trial documents quickly and efficiently. The platform provides features such as generating precise invalidity or evidence-of-use claim charts in minutes, element-by-element claim charts with citations to prior art and product documentation, and claim construction generated from the patent, prosecution history, and family. The platform also includes a search function to find patent documents, non-patent literature, and products in real time. Additionally, it offers drafting tools for winning arguments, including pre-formatted templates and a collaborative workspace for teams to refine documents together in real time. &AI is designed for patent litigators and aims to help them win cases faster with trial-ready analysis. The platform is backed by world-class investors and has advisory board members who are leading voices in patent litigation and strategy. The platform prioritizes security and confidentiality, with enterprise-grade security and compliance built for law, SOC 2 Type II certified, and strict zero data retention policies. Overall, &AI aims to scale patent expertise and help patent litigators craft strong work product quickly and efficiently.
Criteria breakdown
- Automated claim chart generation
- AI-assisted prior art search
- Patent and file history analysis
- Mapping of claims to accused products
- Export to litigation-ready document formats
- Collaboration tools for legal teams
Apify
Web scraping and automation platform for extracting live data to power AI models and agents.
Apify is a cloud platform for collecting and processing web data at scale. Developers can choose from thousands of pre-built scrapers (called Actors) in its marketplace or build custom ones in Python or JavaScript, then run them on managed infrastructure with proxies, scheduling, and storage included. The platform is widely used to feed AI workflows, from gathering training datasets and powering retrieval-augmented generation to enabling autonomous agents that browse the live web. Outputs can be exported as JSON, CSV, or Excel, or pushed directly into databases, vector stores, and LLM pipelines through APIs and integrations with tools like LangChain, LlamaIndex, and n8n. Apify offers a free tier with monthly platform credits, plus paid plans that scale with usage, making it accessible for both individual developers and enterprise teams handling large extraction jobs.
Criteria breakdown
- Marketplace of pre-built scraping Actors
- Managed proxy rotation and anti-blocking
- Cloud runtime with scheduling and monitoring
- Dataset, key-value, and request queue storage
- REST API and SDKs for Python and Node.js
- Integrations with LangChain, LlamaIndex, and Zapier
Firecrawl AI
API that turns entire websites into clean, LLM-ready markdown or structured data.
Firecrawl AI is a web scraping and crawling platform built specifically for AI workflows. With a single API call, it can traverse a full website, render JavaScript-heavy pages, and return clean markdown, HTML, or structured JSON ready to feed into language models, RAG pipelines, or vector databases. The service handles the tedious parts of large-scale extraction, including proxy rotation, rate limiting, dynamic content, and content cleaning. Developers can target single URLs, crawl entire domains, or extract specific fields using schema-based prompts, making it useful for building knowledge bases, training datasets, and AI agents that need fresh web context.
Criteria breakdown
- Full-site crawling with one endpoint
- Markdown, HTML, and screenshot output formats
- Schema-based structured data extraction
- JavaScript rendering and anti-bot handling
- SDKs for Python, Node, and integrations with LangChain and LlamaIndex
- Self-hostable open-source version