Marketing tooling assumes a marketing team: campaigns, funnels, personas, attribution. If you are one engineer who also has to tell people the product exists, almost none of it applies.
Here is the short version of what is worth having.
1. A record of what you built
You already have it. Git history is the highest quality raw material available to a developer, and almost nobody uses it for anything except archaeology.
The gap is not the record, it is that nothing turns it into something you can write from. That is the problem Noomachy DevTracker exists to solve, and if you would rather not use a tool for it, a weekly git log --since="1 week ago" --pretty=format:"%ad %s" into a scratch file does most of the job.
2. Somewhere to put half-formed things
A single text file. Not Notion, not a database, not a tagging system.
The purpose is to catch the thought you had in the shower before it evaporates. Anything with structure adds friction at exactly the moment when friction kills the capture.
3. Two places to post
Pick two. For most developers that is LinkedIn and X, occasionally a blog.
Posting to four platforms badly is worse than two platforms consistently, and cross-posting identical text to both is visible and slightly embarrassing. The formats are genuinely different: X rewards a single sharp observation, LinkedIn rewards a short narrative with a beginning and an end.
4. A place to publish long form
Your own domain. Not Medium, not a platform that can change its rules.
This is the only piece of the stack that compounds. A post on X is gone in two days. A page you own accrues links and search traffic for years.
Next.js with markdown files in a directory is enough. This blog is exactly that: a folder of .md, a reader that parses frontmatter, two routes.
5. Basic analytics
You need to know two things: which posts brought people to the site, and which pages they stayed on.
Any privacy-respecting analytics tool does this. Do not build a dashboard. Do not track conversion funnels. At your scale the numbers are too small for anything sophisticated to mean anything.
What to skip
Scheduling tools, unless you already post consistently. Scheduling is a solution to "I have too much content", which is not your problem.
CRMs. You have no pipeline yet.
Newsletter platforms, until you have something to send. Collecting emails you never mail is worse than not collecting them.
Engagement automation. Auto-likes, auto-comments, auto-DMs. Immediately recognisable and actively negative.
SEO suites. For a small site the advice reduces to: write about specific things, use headings, do not publish thin pages. You do not need a tool to tell you that.
The actual constraint
None of this is a tooling problem. The constraint is that you have limited attention and it is mostly committed to building.
Which means the right stack is the one that takes the least attention per post. Everything above is chosen on that basis and nothing else.