velyan

I’m mapping real founder hacks that protect focus on highest-leverage work.

Looking for concrete, copyable details: • Capture: How do you grab every commitment in-flow (keyboard macros, Raycast/Alfred, email rules)? • Triage → Focus: Your scoring/heuristic (e.g., unblock_others > revenue_this_week > existential). • Follow-ups: How do you guarantee chase-backs (SLA rules, auto-nudges, snooze loops)? • Calendar design: Maker blocks, default 25-min calls, guardrails, “no-meeting” windows. • Automation: Filters, scripts, Shortcuts/Zapier/Make, CRM/email/Slack glue. • Rituals: Daily shutdown, weekly review, “burn down open loops.” • Anti-hacks: What you stopped doing that freed time.

Please reply in this format (paste scripts/gists if you can):

Stage/Team: EA/VA: Y/N Stack (email/PM/CRM/etc): Capture (where + triggers): Prioritization heuristic (exact rule): Follow-up system (tools + SLAs): Automation hacks (scripts/workflows): Weekly review ritual (steps, minutes): Biggest recent miss & fix: One hack that saved ≥5h/week:

I see this in my own workflow all the time: • Exporting a Notion page to Google Docs/Word because reviewers insist on .docx, then re‑attaching a PDF for email. • Copy‑pasting decisions from Slack into Jira/Asana or summarising chat threads into Notion so context isn’t lost. • Manually downloading receipts from emails and entering them into a spreadsheet at month‑end. • Digging through Gmail/Calendar/Docs/Slack to prep for a meeting because context is scattered everywhere.

I’m curious how others handle this. Do you use automations (Zapier/Apple Shortcuts/etc.), or do you just live with the friction? Does app overload and constant format conversion slow you down, or is it a minor annoyance?

Would an on‑device assistant that can route files (e.g., GDocs→PDF→draft email), sync context to your notes, and prep meeting packets actually help, or is the problem overstated?

Would love to hear concrete examples of what’s painful, what’s working, and what you wish existed.

Hi HN!

I’m excited to share Collate, a macOS app I built that lets you chat with PDFs, generate summaries, and read documents—all offline. It runs entirely on-device using Llama 3.2 quantized models, ensuring privacy without compromising on capability.

Features: • Ask Questions: Interact with PDFs to extract key insights or answers without manual searching.

• Summarize: Generate concise summaries from long documents in seconds.

• Privacy-First: Built with on-device processing—your data never leaves your machine.

• Offline-Ready: Works without an internet connection, powered by efficient quantized models.

Under the Hood:

• Runs Llama 3.2 quantized modelels.

• Optimized for macOS to provide a seamless offline experience.

The app is free and available on the macOS App Store.

Demo: https://collate.one/get-started#features

Download: https://apps.apple.com/us/app/collateai/id6447429913

I’d love your thoughts on features, performance, or ideas to make this better. Thanks for taking a look!

Let me know if you’d like further refinements!

Curious what other people are learning these days

Why are they your favourite?

What learning resources would you recommend for graph theory and why ate they good? I have a CS degree, but need a deeper dive and examples of practical applications. Have you recently read articles, books, watched youtube videos. If you have a list of resources you have personally gone through that would be ideal! Thanks