Skip to main content

Claude/ChatGPT Prompt to Turn a Technical Win into a LinkedIn Post

Turn a dry technical win into an engaging, under-200-word LinkedIn post with a scroll-stopping hook that recruiters and engineering peers actually read.

Fill in the placeholders

Edit the values, then copy your finished prompt.

Your Prompt
prompt.txt
You are a senior engineer who writes well. Produce a finished LinkedIn post, not bullet notes.

Context:
- The win: Cut API p95 latency from 1.2s to 240ms by replacing an N+1 ORM call with a single SQL window function.
- Audience: engineering peers and technical recruiters
- Voice: direct, humble, no buzzwords

Deliver:
1) A first-line hook built on a surprising number or counter-intuitive claim.
2) Three to five short paragraphs with line breaks for scannability.
3) Concrete numbers (time saved, cost reduced, latency cut, users helped).
4) One lesson paragraph that generalises the insight beyond this case.
5) A closing question that invites comments.

Output as a single post under 200 words, no corporate speak, at most one or two emojis.

What this prompt does

This prompt turns a buried technical achievement into a finished, under-200-word LinkedIn post. It assigns the AI the role of a senior engineer who writes well, then forces a specific shape: a hook built on a surprising number, three to five scannable paragraphs, concrete metrics, one generalisable lesson, and a closing question. You feed it three things via [win_description], [audience], and [voice], and it returns a post you could paste straight into the composer.

The structure works because it separates the raw win from the storytelling. [win_description] carries the facts (the latency cut, the cost saved), [audience] aims the framing at the right readers, and [voice] keeps the tone from drifting into corporate speak. By insisting on a real number in the first line and capping the whole thing under 200 words, the prompt fights the two failure modes of engineering posts: vague "improved performance" claims and walls of unbroken text.

When to use it

  • You shipped something measurable and want to share it without sounding like a press release.
  • A win is sitting in a commit message or a Slack thread and you keep meaning to write it up.
  • You are building a professional presence for recruiters and engineering peers at the same time.
  • You want a consistent posting voice instead of reinventing the tone each time.
  • You need a draft fast and would rather edit than stare at a blank box.
  • You want a closing question that actually invites replies rather than a flat sign-off.

Example output

You get a single, ready-to-post block: one punchy first line carrying a real number, a few short paragraphs separated by line breaks, a lesson paragraph that lifts the insight beyond your specific case, and a question at the end. It stays under 200 words, uses at most one or two emojis, and avoids buzzwords. It is a finished post, not an outline or bullet notes.

Pro tips

  • Put one concrete metric directly in [win_description] — the hook is only as strong as the number behind it, so "1.2s to 240ms" beats "made it faster."
  • Tune [audience] deliberately: a post for technical recruiters lands differently than one for senior peers, and the framing shifts with it.
  • Use [voice] to stop the model from inflating things; values like "direct, humble, no buzzwords" keep it honest.
  • If the first hook feels generic, re-run with a more counter-intuitive angle on the same win rather than rewording the whole post.
  • Keep the lesson paragraph general but earned — it should teach something even to readers who do not care about your specific stack.
  • Trim ruthlessly toward the 200-word cap; the constraint is what makes it scannable, so resist re-adding detail the post does not need.

Frequently Asked Questions

Will the post actually stay under 200 words?
The prompt instructs the model to output a single post under 200 words, and most models respect that. Always check the final count yourself, since models occasionally run slightly over, and trim the lesson paragraph first if it does.
Does it work for non-engineering wins?
It is tuned for technical wins and asks for engineering-style metrics like latency or cost, but you can adapt `[win_description]` for adjacent work. The hook-and-lesson structure still holds for any achievement that has a real number behind it.
What if my win does not have a clean number?
The prompt leans hard on a concrete metric in the hook, so a win with no measurable result will produce a weaker post. Find any honest number first — time saved, queries reduced, users helped — before running it, or the output reads vague.
Can I control the tone so it does not sound like marketing?
Yes, that is what `[voice]` is for. Setting it to something like "direct, humble, no buzzwords" tells the model to drop corporate speak, and the prompt already caps emojis at one or two to keep it from feeling salesy.
Engr Mejba Ahmed

Need this built for real?

Engr Mejba Ahmed

AI Developer · Software Engineer

I'm Mejba — I design and ship production AI systems, automations, and full-stack apps. If you want this turned into a working solution for your team, let's talk.

More in AI Writing for Developers

Engr Mejba Ahmed

Engr Mejba Ahmed

AI assistant · trained on my work

👋

Hey there!

Quick Actions

WhatsApp Direct line to me

Chat on WhatsApp

+880 1723 741224 · Replies within the hour on working days

Popular Questions

Engr Mejba Ahmed is connected
Engr Mejba Ahmed is typing...
Engr Mejba Ahmed avatar

✉ Want me to follow up? Drop your email

Engr Mejba Ahmed avatar

📞 Connect Directly

Choose how you'd like to reach me

WhatsApp

+880 1723 741224

Email

mejba.13@gmail.com

✓ Details sent! I'll get back to you shortly.

Powered by OpenAI

335+

Blog Posts

25

AI Courses

63

Projects

Services & Expertise

Pricing & Process

Learning & Resources

Connect & Support