LinkedIn

📘 Learn LinkedIn from zero

Start from zero. Imagine every professional in the world is a shop on one giant street. A resume is a paper flyer you hand to one shopkeeper when you apply for a job — it disappears into a drawer. LinkedIn is your shopfront window on that street: it is always open, anyone can walk past, and — crucially — there is a search desk where recruiters type keywords like "frontend developer React" and get a ranked list of windows to visit. Your job is to make your window findable and worth stopping at.

A LinkedIn profile is just structured data about you, broken into sections: a Headline (one line under your name), an About summary, Experience, Skills, and a Featured area where you pin links and media. Recruiters search across these fields, so the words you choose are a primary input to whether you surface in their results — keywords are not the only signal (activity, connections, and profile completeness matter too), but they are the one you fully control.

Worked example. Suppose you just built a weather app in React. The weak window says: Headline: "Student." Featured: empty. Nobody searching finds you, and nobody who lands stays. The strong window: Headline: "Frontend Developer | React, TypeScript | Building accessible web apps." About: two sentences on what you do and what you are looking for. Featured: a pinned link to the live app plus its GitHub repo, with a screenshot as the thumbnail. Now a recruiter searching "React" is more likely to find you, reads a clear value line, and clicks straight through to proof of work.

The single key insight: LinkedIn is a search-and-discovery surface, not a document — optimize for the keywords recruiters search and the proof they click, not for prose nobody reads end to end.

✨ Added by the guide to build intuition — not from the source course.

🎯 Guided practice

1. Easy — Rewrite a dead headline. Given headline: "Recent CS Graduate." Step 1: identify what a recruiter would type to find this person — they search by role and tech, not by status. "Graduate" is a keyword nobody searches for when sourcing candidates. Step 2: use the pattern [Role] | [Key Skills] | [Value or Focus]. Step 3: fill it: "Backend Developer | Python, PostgreSQL, AWS | Building scalable APIs." Why it works: the role and each skill are now searchable tokens, and the value clause gives a human reason to click. Core pattern: every field is a search index — spend its characters on terms people query, not on labels that describe your current state.
2. Medium — Use ChatGPT to draft an About section, then de-risk it. Task: turn three facts (built a weather app in React; led a 4-person hackathon team; want a frontend role) into an About summary. Step 1 (prompt the model): give it the facts plus constraints — "Write a 3-sentence LinkedIn About in first person, no buzzwords like 'passionate' or 'synergy', and only claim what I gave you." Constraining the input is what separates usable output from generic filler. Step 2 (verify, do not trust): read every sentence and ask "can I defend this in an interview?" Delete anything you cannot back up — this is the same discipline as never shipping code you have not read. Step 3 (inject keywords + proof): ensure "React" and "frontend" survive the edit (they are search tokens), and pair the About with a Featured link to the actual app so the claim is verifiable. Step 4 (humanize): rewrite one sentence in your own voice so it does not read as machine output. Core pattern: AI is a draft accelerator, not an author — you supply the facts and constraints, the model supplies structure, and you own the final verification and proof links.
3. Hard — Wire up Featured media and a maintenance cadence so the profile stays trustworthy. Task: take the React weather app and make it a clickable, evidenced Featured item, then define how you keep the whole profile from going stale. Step 1 (add the right media, link don't host): in Featured, add the live app URL and the GitHub repo URL as separate items — LinkedIn auto-fetches a link preview (title, description, image), so set a clear repo description and a social-preview image on the GitHub side so the thumbnail is not blank. Treat Featured as a curated index of proof, not a dumping ground: 2–4 items, strongest first. Step 2 (caption for the skimmer): override each item's auto-title with a one-line "what this is and what it shows" caption — a recruiter scans captions, not READMEs. Step 3 (close the keyword loop): confirm any tech named in the caption (React, TypeScript) also appears in your Skills section, since cross-field reinforcement is what a keyword search rewards. Step 4 (define the update trigger, not a calendar): maintenance is event-driven, not periodic — set a personal rule: "every time I ship something or change roles, update Headline + Featured + Experience in one batched pass." This keeps the cheap O(1) edits aligned with reality and prevents the stale-profile pitfall. Step 5 (guard the "Open to Work" signal): only raise the Open to Work flag when the profile is current, and lower it when you are not actively looking — a live flag on a stale profile reads worse than no flag. Core pattern: Featured turns claims into clickable evidence, and an event-triggered update rule (ship → update) keeps that evidence and your availability signal honest with near-zero ongoing effort.

✨ Added by the guide — work these before the full problem set.