Monday, 8:47 a.m.
Seven Slack DMs you haven't opened. Three 1:1s stacked before lunch. A half-written planning doc from Friday. A decision from last quarter you need to remember, buried in a thread you can no longer search. A promotion packet open in three browser tabs. Two incident follow-ups you said you'd reply to "first thing Monday."
It's 8:47 a.m.
You have not had coffee.
This is the job. The job isn't writing code. It isn't even running standups. The engineering manager's actual job is carrying context between decisions without losing it. And your brain, with respect, is a terrible database.
🌀 Read Issue 001 here: https://read.englead.ai/p/em-operating-system-ai-native
What I built
I built a Second Brain for this. I call it The EM Operating System. It took one Saturday afternoon. It runs local, costs nothing beyond a Google account, and lives on top of four things you can install with five terminal commands.
The four pieces:
1. Obsidian
Obsidian is a plain markdown editor that treats your notes as files in a folder on your laptop. Not a SaaS. Not a database. Just files.
Why it matters for an EM: everything you write about work (1:1 notes, architecture decisions, hiring feedback, retro threads) is permanently yours. Obsidian shuts down tomorrow? Your notes still open in any text editor. That reversibility is why senior engineers can put real work context into it without worrying about vendor risk.
It also renders markdown beautifully, supports backlinks between notes, and has a mobile app. Free for personal use.
2. Claude Code
Claude Code is an AI agent that runs in your terminal and reads the files in the directory you point it at. Not a chat app. Not a sidebar. An agent that writes, edits, organises, and links files on your behalf.
Why it matters for an EM: you point Claude at your ~/brain folder once. From then on, every capture you throw at it (a voice memo transcript, a Slack screenshot, three sentences from a 1:1) gets filed in the right place, with frontmatter, with backlinks to the people and projects it touches. The filing stops being your job.
The unlock isn't "AI writes my emails." The unlock is "AI carries the connective tissue between my 1:1s, roadmap reviews, incident notes, and hiring loops, so I don't drop threads."
3. Google Drive + rclone
rclone is a command-line tool that syncs any folder to any cloud storage. Install it once. Point it at your ~/brain folder. It now syncs to Google Drive every 15 minutes.
Why it matters for an EM: you work on a laptop at the office, a different laptop at home, a phone on the commute. Obsidian Sync costs $10/month. Dropbox and iCloud both have weird edge cases with Obsidian's file locks. rclone + Google Drive is free, reliable, and uses storage you already have.
And because Obsidian operates on files, not a proprietary cloud database, rclone sync is a real-time replica. Not a backup. A working mirror.
4. PARA
PARA is a folder method from Tiago Forte: Projects / Areas / Resources / Archive. Four folders. Everything filed into exactly one of them.
Why it matters for an EM: it's action-first, not archive-first. Projects have a finish line. Areas are your ongoing responsibilities (the team, the role, 1:1s with your manager). Resources are reference material. Archive is done-or-abandoned. Anything else goes in a fifth folder, Inbox, and gets processed weekly.
No tag taxonomy. No nested subfolders. No database schema to memorise. Five folders, one rule: new things land in Inbox; old things move out.
~/brain
├── 0 - Inbox ← anything you capture ends up here first
├── 1 - Projects ← active work with a finish line
├── 2 - Areas ← ongoing standards (the team, the role)
├── 3 - Resources ← reference material, people, captures
└── 4 - Archive ← done, paused, or moved on
How it actually runs on a Monday
Back to 8:47 a.m. Your report mentions a decision from two weeks ago in the opening minute of your 1:1.
Old world: you apologise, open five Slack search tabs, scroll through a thread, fail to find it, promise to circle back.
New world: you type a single line into Claude Code. "What did we decide about the retry policy on payments-service in the last three weeks?" It reads your vault, finds the ADR, finds the 1:1 note where the decision was referenced, finds the Slack screenshot you captured, and gives you the summary plus links. Thirty seconds. You're back in the conversation.
Same capability, different layer. The decision was always there. You just stopped being the index.
Why "operating system"
I call it an operating system because that's what it is. Not an app. Not a note-taking tool. An opinionated stack of primitives that together run the EM's knowledge work the way the OS on your laptop runs your applications.
- Obsidian is the filesystem.
- Claude Code is the process scheduler (it decides which files to read, write, and link based on what you ask).
-
rclone+ Google Drive is the network layer. - PARA is the directory structure.
You already understand operating systems. This one just runs on your knowledge work instead of your apps.
The full setup
I wrote Issue 001 of EngLead AI as the complete walkthrough: every install command, every copy-paste prompt, the rclone sync config, the para-capture Claude Code skill that turns the agent into your personal librarian, and the Obsidian settings that make it all feel native.
About 45 minutes end to end. Zero prior experience with Obsidian, PARA, or Claude Code required. Mac only for now.
🌀 Read Issue 001 here: https://read.englead.ai/p/em-operating-system-ai-native
Writing The AI-Native Engineering Leader: on combining AI with existing technology to manage and operate teams at scale.
Subscribe at read.englead.ai for the weekly issue.
Top comments (14)
The 'brain is a terrible database' line is a hard truth. I’m currently mid-build on Commerza (a framework-less ecommerce engine), and I learned this the expensive way.
I hit a major refactoring disaster recently where I lost 40% of my backend logic. What actually saved the project wasn't just my code backups, but the fact that I had documented my 'security primitives' (SQL transactions, auth logic, etc.) separately. It turned a total loss into a 9-hour reconstruction job instead of a 1-month failure. I’m definitely looking into the PARA folder structure for my next project to keep those architecture decisions even more accessible.
Solid approach, love the mix with Obsidian and GDrive
Yes this is applicable for any usecase not just managers but also engineers / builders.
This is similar to my own! I've recently added hooks and started logging the conversation back and fourth and I inject memories using hooks and a vector db (if I have memory mode on) into the prompt itself. I have ADHD and memory issues, so I like to be reminded of what happened in the past, but it actually seems to improve a lot of answers since so much of what I discuss with AI is related.
I did add a scripts and journal folder, both for work and both shared with other projects. I have a wiki folder, too, where all the processed files live. That's a pretty big strike against the methodology, but I think it's important that we all change the system to fit our own needs. My wiki folder is the same information found in other areas, just written from a different, technical perspective vs a lot of opinion and emotion found elsewhere. That's how it's broken up in my head, so that's how it's broken up in my second brain.
Where can I see all of the steps to configure the shared brain? Do I need paid subscription with Claude?
based on the look of it wou'll need a Max or at least a Pro plan
Mine requires the smaller of the two max plans, and it's pretty similar. This doesn't require top models for everything, though. A lot of processing can be done with haiku if you write tight enough prompts, but that's not always easy when exploring new concepts.
My first version of this used opus to write the prompts and md files, and haiku to actually execute those prompts. Then had sonnet fix things along the way. I used regular Claude app for the design phase, Claude Code for the rest. I don't have the best frame of reference (I'm spoiled), but I think that would have worked with Pro usage levels.
Not great but still like the tool and how it works.
This all originated (the fad, not the concept) with a tweet by Andaarej Karpathy, a respected voice in the AI space. This is the main gist of his setup, which is pretty much what's being described in the post above.
Since this post, there have been a BUNCH of youtube videos and internet tutorials. Best bet is searching "Obsidian Second Brain" and finding a resource that speaks to you personally. It does require a subscription, but it ranges in usage depending on how often you use it and how you use it. I use mine for everything, coding included, so I need a Max subscription, but someone using it occasionally might only need a Pro subscription. It'd be really tough on a free subscription.
I have done something similar for my personal project tracking.
This is really good idea I have been thinking how to organize all my vibe coding ideas.
Yes this is applicable for any usecase not just managers but also engineers / builders.
The "brain as terrible database" framing is a great reframe. The adjacent problem I keep running into: most second-brain systems are great at write and terrible at recall. I've had the most luck with a flat daily-notes markdown tree + grep + a small semantic search layer on top — the structure stays dumb enough that tools (and agents) can read it, but retrieval still works because of embeddings. Heavy folder hierarchies tend to become the place notes go to die.
i've been using a physical notebook for morning dumps and it's wild how much it reduces that mental cache pressure. no notifications no context switching. feels like defragging my brain before the day even starts