Grammarly / ai-tools

From Idea to Demo in Two Days: Inside Superhuman’s 2025 Global Hackathon (opens in new tab)

Superhuman’s 2025 hackathon brought together nearly 500 employees to prototype innovative product features by leveraging cutting-edge AI coding tools like Claude Code and Cursor. By integrating AI-driven agents and keyboard-centric workflows, teams demonstrated how rapid experimentation can bridge functional gaps across mail, documentation, and collaboration platforms. The event highlighted a significant shift toward "vibe-coding" and accessible development, where cross-functional teams and non-engineers could ship functional MVPs in just 48 hours. ## Superhuman Command Everywhere (SCE) * This project extends the Superhuman Mail Command Center to the browser, allowing users to trigger Grammarly features, set reminders, and snooze items from any web page. * The tool enables keyboard-only navigation for AI agents; for example, users navigate Grammarly’s Proofreader cards using "J" and "K" and accept or dismiss suggestions with "E" and "D." * Developers used AI tools to quickly interpret an unfamiliar codebase, allowing engineers without frontend expertise to "vibe-code" a working MVP within a few hours. ## Whiteboarding in Coda * This feature introduces a native canvas within Coda documents where users can draw freely, add shapes, and import images for brainstorming and diagramming. * The prototype includes an AI diagramming tool that generates editable visual versions of diagrams based on plain-text descriptions. * Built by a solo team member with no formal coding background, the project utilized Claude Code and Cursor to focus on UX refinement and smooth interactions rather than just technical functionality. ## Superhuman Listening * This system centralizes fragmented customer feedback from tools like Gong, Salesforce, and Zendesk into a single, queryable source of truth. * By linking unstructured data to product roadmaps in Coda, the tool helps sales engineers and product managers determine if specific customer feedback is already being addressed. * Technical challenges included using LLM APIs to extract urgency and sentiment, though the team noted the difficulty of filtering "noise" from high-volume sources like Zendesk tickets. ## Inclusive Language Agent * Developed by a team of linguists, this agent identifies non-inclusive phrasing or unconscious bias in professional writing. * The goal is to provide real-time suggestions that improve workplace culture and customer trust by making word choices more inclusive and intentional. The results of this hackathon suggest that AI-assisted development tools are significantly lowering the barrier to entry for complex product builds. For organizations aiming to accelerate innovation, encouraging "maker" identities across all departments and utilizing AI to bridge technical skill gaps can surface high-value solutions that traditional product cycles might miss.

Agentic AI vs. generative AI: What’s the Difference and When to Use Each (opens in new tab)

While generative AI focuses on creating content like text and images through prompt-based prediction, agentic AI represents a shift toward autonomous goal achievement and execution. By combining the creative output of large language models with a continuous loop of perception and action, these technologies allow users to move from simply generating drafts to managing complex, multi-step workflows. Ultimately, the two systems are most effective when used together, with one providing the ideas and the other handling the coordination and follow-through. ### Distinguishing Creative Output from Autonomous Agency * Generative AI functions as a responder that produces new content—such as text, code, or visuals—by predicting the most likely next "token" or piece of data based on a user’s prompt. * Agentic AI possesses "agency," meaning it can take a high-level goal (e.g., "prepare a client kickoff") and determine the necessary steps to achieve it with minimal guidance. * While tools like Midjourney or GitHub Copilot focus on the immediate delivery of a specific creative asset, agentic systems act as proactive partners that can use external tools, manage schedules, and make independent decisions. ### The Underlying Mechanics of Prediction and Action * Generative models rely on Large Language Models (LLMs) trained on massive datasets to identify patterns and chain together original sequences of information. * Agentic systems operate on a "perceive, plan, act, and learn" loop, where the AI gathers context from its environment, executes tasks across different applications, and adjusts its strategy based on the results. * The generative process is typically a direct path from input to output, whereas the agentic process is iterative, allowing the system to adapt to changes and feedback in real-time. ### Practical Applications in Content and Workflow Management * Generative use cases include transforming rough bullet points into polished emails, summarizing long documents into flashcards, and adjusting the tone of a message to be more professional. * Agentic use cases involve higher-level orchestration, such as monitoring document revisions, consolidating feedback from multiple stakeholders, and automatically sending follow-up reminders. * In a project management context, an agentic system can draft a project plan, identify owners for specific tasks, and update timelines as milestones are met or missed. ### Navigating Technical and Operational Limitations * Generative AI is susceptible to "hallucinations" because it prioritizes probabilistic output over factual reasoning or logic. * Agentic AI introduces complexity regarding security and permissions, as the system needs authorized access to various apps and tools to perform actions on a user's behalf. * Current agentic systems still require human oversight for critical decision-making to ensure that autonomous actions align with the user's intent and organizational standards. To maximize efficiency, you should utilize generative AI for the creative phases of a project—such as brainstorming and drafting—while delegating administrative overhead and coordination to agentic AI. As these technologies continue to converge, the focus of AI utility is shifting from the volume of content produced to the successful execution of complex, real-world results.