Rewind

We're the first productivity system built around disruption, not planning. Designed for ADHD brains, but for improving productivity for everyone.

Comment

Inspiration

6.1 million adults in the U.S. are diagnosed with ADHD.

But diagnosis isn’t the problem. Executive dysfunction because of ADHD means that even when you know exactly what you need to do, starting feels impossible. The planning isn’t the problem, it’s truly when the plan breaks.

Two members of our team live with ADHD. We’ve tried every productivity system: Notion databases, time-blocking on Google Calendar, even gamified to-do apps. They all share the same fatal flaw: they help you plan, but they abandon you the moment something goes wrong.

A meeting runs 15 minutes over. You space out for 10. By the time you’ve replanned, 45 minutes are gone, and your momentum with it.

That's the exact moment we're targeting: the 15 to 30 minutes after a disruption occurs, when the plan breaks and the user is silently stuck because they have no idea what to do next.

What It Does

The moment we realized this wasn't a willpower problem, but a working memory system, an obvious solution came to mind: what if we could recreate an operating system? What if we took inspiration from the way our computers batched and scheduled tasks and this system for us?

When processes compete for limited CPU time and interrupts fire unexpectedly, operating systems don’t panic. They rely on scheduling algorithms:

  • Long-Term Schedulers plan allocation.
  • Medium-Term Schedulers handle swaps.
  • Short-Term Schedulers decide what runs right now.

They don’t delete processes when something changes, rather, they're able to adapt and reschedule them.

Rewind applies this model to human productivity. We create an operating system for your life: a completely new way to look at productivity for the ADHD community.

All the workflows we created maps to a specific ADHD challenge documented in clinical research:

  • Disruption Detection → Disruption Paralysis
  • Auto-Rescheduling → Cognitive Overload from Replanning
  • Estimation Correction → Time Blindness
  • Energy-Aware Scheduling → Working Memory Limitations

We achieved real-time response to life disruptions. From the moment a meeting runs over to a nudge appearing on screen: under 3 seconds. That's fast enough to catch someone before they spiral into doom-scrolling.

Our Agent Framework

Our system has six agents, all sharing context and communicating with each other.

  1. Context Sentinel: Monitors your Google Calendar, Gmail, and Slack (among other integrations) continuously. We use this metadata to see when a disruption occurs, and the moment reality diverges from your plan.

  2. Disruption Detector: Receives context changes and categorizes severity: minor (meeting ran 5 min over), major (meeting cancelled), or critical (urgent Slack from your manager).

  3. Scheduler Kernel: Runs our three-tier scheduling engine (Long-Term Scheduling/Mid-Term Scheduling/Short-Term Scheduling) inspired by operating system process schedulers, coordinating all schedule changes.

  4. Energy Monitor We use research-backed data on circadian patterns and metadata on how you complete your tasks to identify the best current task for you. Our goal? Never assign a task your brain can't handle right now.

  5. Profiler Agent Studies six behavioral patterns: your actual peak hours for task completion (not when you think you're productive), your estimation bias (how much you underestimate the time to complete tasks), your energy curve, and more. Gets smarter with every interaction.

  6. GhostWorker When being proactive means we can save you from disruption: GhostWorker drafts emails, Slack replies, and meeting reschedules for you. You approve with one tap.

The core insight: every productivity tool helps you make a plan. Rewind is the only one that helps when the plan breaks, because we know what works for you.

Our OS-Inspired Scheduler

We actually implemented core hardware principles, now abstracted to productivity: Modified Multilevel Feedback Queue (MLFQ), bin-packing, and swap-in and swap-out.

  • LTS (Long-Term Scheduler): Daily planning. Scores tasks using a weighted algorithm (40% deadline urgency, 30% priority, 15% peak-hour alignment, 15% duration efficiency) and bin-packs them into your available hours with estimation bias correction.
  • MTS (Medium-Term Scheduler): Disruption recovery. When time is freed or lost, MTS runs swap-in/swap-out operations between your active schedule and backlog buffer. Energy-aware: tasks exceeding your current capacity are never scheduled.
  • STS (Short-Term Scheduler): Task ordering. Implements a Modified Multilevel Feedback Queue (MLFQ) with four priority levels (P0-P3), supporting preemption for urgent tasks while respecting energy constraints.

Tech Stack

Backend:

  • Python with FastAPI for the core server
  • uAgents (Fetch.ai) for our decentralized agent architecture—each agent runs on its own port with deterministic addressing
  • Composio MCP as our multi-platform orchestration layer for Google Calendar, Gmail, Slack, and LinkedIn integrations
  • Redis for state management, caching, and pub/sub messaging
  • WebSocket for real-time frontend updates (sub-100ms latency)

Frontend:

  • Next.js 16 with React 19
  • TypeScript with strict typing
  • Tailwind CSS 4 with a clean, YC-inspired light theme
  • Custom state management via Zustand-pattern hooks
  • WebSocket client for real-time schedule updates

Agent Communication:

  • Typed message protocols (ContextChangeEvent, DisruptionEvent, EnergyQuery)
  • Address-based routing via seed-derived deterministic addresses
  • Redis pub/sub for GhostWorker draft approval workflows

What we learned

We learned realized that research-first design produces better systems than feature-first design.
When we started by listing "cool features," we built things nobody needed. When we started by identifying the specific cognitive failure points (disruption paralysis, cognitive overload, time blindness), the features emerged naturally. The research constrained us in productive ways, and means we built a project we're incredibly proud of. It's a tool changing how millions of individuals with ADHD navigate their work.

Built With

Share this project:

Updates