Nicheloom visitors: 14,341 all-time · 8,740 in the last 7 days

The opportunity tracker for new startups.

Today: 1,383 launchs · Niche of the day: document and PDF processing tools · See full report →

Drifty for Windows

Private AI time tracker that learns how you work

Get picks like this daily. The day's top launches, AI/tech news, and a weekly opportunity spotlight — straight to your inbox.

This is 1 of 49 launches in personal habit and activity tracking apps — see how it stacks up on momentum and crowding →

111 other launches read as similar to this one →

Details

External ID
1278531
Source
PH
Company
—
Product
Drifty for Windows
Website domain
producthunt.com
Launched
Oct. 13, 2026
Cohort
—
Upvotes
13
Upvotes percentile
0.95
Tags
Productivity, Time Tracking

Description

Drifty is a private AI time tracker that learns how you work. It tracks your day automatically, with no timers or manual entries, and separates focus from distraction based on context. The more you use it, the better it knows you, as your work profile, what you did before and after each activity, and every correction you make shape how it reads your day. Local mode is completely free, your data never leaves your device, and no screenshots or recordings are ever stored.

Enrichment

Niche
personal habit and activity tracking apps
Vertical
Horizontal
Function
Analytics & BI
Audience
Prosumer
AI stance
AI-native
Project type
Commercial product
Normalized one-liner
ai-powered time tracker for windows users
Manually corrected
False

Could you build this?

Partial Building a standard desktop UI with background window/process tracking is straightforward, but doing accurate, low-overhead, on-device context learning and classification requires non-trivial local ML and OS hook optimization.

What it would actually take: A real implementation requires a native Windows service or Electron/Tauri app utilizing Windows UI Automation and Win32 APIs (SetWinEventHook, GetGUIThreadInfo) to continuously poll active windows and titles. It needs an embedded vector store or lightweight local NLP model (e.g., ONNX runtime with quantized embeddings) to classify context offline without draining battery. The hard part is fine-tuning an adaptive personal activity model that generalizes reliably without leaking sensitive context.

Competitors

Other products that read as similar to this one — 111 launches clear the similarity bar, closest 8 shown.

Attention rank: #8 of 112 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).

Launched 211 days after the earliest competitor.

Other launches for this product