DEV Community: Rohit Itagi

GET OUT — An AI Assistant That Wants You to Stop Using It

Rohit Itagi — Tue, 06 Oct 2026 18:55:02 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass

What I Built

GET OUT is a local AI app designed to help people spend less time on their screens and more time in the real world.

The idea is simple:

The best AI session is the shortest session.

Instead of keeping users inside a chatbot, GET OUT gives them one real-world mission and then gets out of the way.

The user chooses:

⏱️ Time: 10 minutes, 30 minutes, 1 hour, or 2+ hours
⚡ Energy: Low, Normal, Active, Social, Quiet, or Surprise Me

The local AI then generates exactly one physical mission based on those choices.

For example:

You have 30 minutes. Explore a familiar route you normally ignore. Find one unusual detail you've never noticed before and investigate it for a few minutes.

PHONE DOWN. GO.

There is no recommendation feed, endless conversation, or AI companion.

The flow is:

USER → 2 CHOICES → LOCAL AI → ONE MISSION → PHONE DOWN → GO OUTSIDE

GET OUT is for people who feel like they are spending too much time on their phone or computer and need a small push to step away.

The goal isn't to maximize AI engagement.

The goal is to make the user leave the app.

Demo

Demo: [Add your demo link here]

The demo shows the complete flow:

Choose how much time you have
Choose your energy level
Get one AI-generated mission
Put the phone down
Go outside

Code

GitHub: [Add your GitHub repository link here]

The project is open source and designed to run locally.

How I Built It

GET OUT uses open-weight AI with local inference.

Tech Stack

Qwen3 4B — open-weight language model
Ollama — local AI inference
FastAPI — backend API
Python — application logic
HTML/CSS/JavaScript — frontend
pytest — automated testing

The architecture is intentionally simple:

User
  ↓
Web UI
  ↓
FastAPI
  ↓
Mission Generator
  ↓
Ollama
  ↓
Qwen3 4B
  ↓
Safety Validation
  ↓
Mission Quality Validation
  ↓
ONE MISSION
  ↓
PHONE DOWN
  ↓
REAL WORLD

The model receives the user's available time and energy level and generates a mission designed around those constraints.

The generated mission is then validated before being shown to the user.

I intentionally avoided building another AI chatbot.

There is no endless conversation, recommendation feed, or AI companion.

AI generates the mission. Then the user leaves the screen.

Why Does Open Innovation Matter?

GET OUT is built around local, open-weight AI because the project does not need a large cloud AI service to work.

Using open AI makes it possible to:

Run the model locally
Avoid sending user context to a third-party AI API
Experiment directly with the model and prompts
Run without a paid inference API
Modify the AI behavior
Experiment with different open-weight models
Keep the project accessible to other developers

But there is also a bigger reason.

Most digital products are optimized for:

more engagement → more sessions → more screen time

GET OUT is designed around the opposite goal:

less AI usage → more real-world time

That creates an interesting product philosophy:

We built an AI assistant whose success is measured by how quickly you stop using it.

The AI is not the destination.

It is the push toward the destination.

Prize Categories

Open-Source Contributor

Stop Searching Your Screenshots. Just Ask.

Rohit Itagi — Sat, 03 Oct 2026 11:15:16 +0000

This is a submission for the

Remember Why — You Saved It for a Reason

What I Built

The Problem

People save screenshots while learning and working because something seems important at the time. Later, those screenshots become difficult to search, and the original context is often forgotten.

I wanted to solve that problem with a simple question:

What if you could ask your own saved screenshots what you were interested in at the time?

Built for a Friend

I built Remember Why for my friend Rakshitha, who regularly saves screenshots while learning and working, but later has trouble remembering what she saved and why it mattered.

The goal was to turn those forgotten screenshots into a searchable personal memory instead of another folder of files.

The Solution

Remember Why is a local-first AI memory assistant that helps people rediscover forgotten context behind saved screenshots.

Instead of searching through hundreds of screenshots by hand, you can ask questions such as:

“What did I save about MCP?”
“What did I save about PyTorch?”
“Why did I save MCP_Architecture.png?”

Remember Why retrieves relevant screenshot memories using semantic search.

For “why did I save this?” questions, it combines temporal proximity with semantic relationships to provide possible context.

It deliberately does not pretend to know the user's true intent. Retrieved OCR, filenames, timestamps, and similarity scores are treated as evidence, while inferred context is clearly labeled.

Demo

Video Demo: https://youtu.be/8UyR7Y0YTQ8

The demo shows:

Asking Remember Why about saved MCP memories.
Retrieving relevant memories using semantic search.
Displaying the original saved screenshot.
Exploring related memories.
Showing possible context.
Observing the LangGraph agent execution through Sentry.

Code

GitHub: https://github.com/Rohith-Itagi7/remember-why.git

How I Built It

Qwen3 4B + Ollama — local open-weight LLM for AI inference.
LangGraph — orchestrates the AI agent and its tool-calling workflow.
MCP Python SDK — exposes screenshot-memory capabilities as tools the agent can use.
Tesseract OCR — extracts text from screenshots locally.
Sentence Transformers (all-MiniLM-L6-v2) — converts screenshot text into semantic embeddings.
FAISS — stores and searches those embeddings for semantic memory retrieval.
Sentry Agent Tracing — provides observability into the actual agent, model, and tool execution.

The core pipeline is local:

Screenshots
    ↓
Tesseract OCR
    ↓
Sentence Transformers
    ↓
FAISS
    ↓
MCP Tools
    ↓
LangGraph Agent
    ↓
Qwen3 4B via Ollama
    ↓
Grounded Response

Why Does Open Innovation Matter?

Privacy: The core memory pipeline runs locally, so personal screenshots do not need to be uploaded to a closed AI API.
Model freedom: Qwen3 4B runs locally through Ollama, and the model can be changed without redesigning the application around a single proprietary API.
Open agent architecture: LangGraph and MCP let me control how the agent discovers and uses memory tools instead of relying on a fixed closed-agent workflow.
Custom retrieval: Sentence Transformers and FAISS let me build and modify the semantic-memory layer myself.
Transparency: The OCR, embeddings, retrieval, MCP tools, and agent behavior can be inspected and modified rather than treated as black boxes.
Cost: Once the open-weight model and local components are installed, the core AI workflow can run locally without per-request AI API costs.
Reproducibility: Others can run and adapt the project using the same open components with their own local data.

Prize Categories

Sentry Agent Tracing

Remember Why uses Sentry Agent Tracing to observe the actual LangGraph agent execution, including model operations, MCP/tool execution, and HTTP operations.

Team Submission

Solo submission — no teammates.