July 20, 2024
As someone who almost never brags, writing an entire blog about my small wins is way out of my comfort zone. But from the countless self-help articles I've read over the year, I guess that's where growth is. Also, if I don't speak of my acheivements, who else will?
[insert 100X🧿]
None of what I've acheived are overnight successes. I'm also almost never the smartest person in the room. Every win has come from blood, sweat and tears(a lot if it). It's come from burning the midnight oil consistently, a strong work ethic and sacrifices you can't even imagine of.
Also lots of prayers.
Starting off with a most recent one.
It was almost two weeks ago when I started researching heads-down about MCP servers and how I could build one for SigNoz(I work here!). So after a lot of research, I started building one. Nothing huge, just a demo MCP server.
The technical writer in me in is urging to explain what MCP is and it's functionalities, but let's press pause on that. For now, if you are a curious soul, I highly recommend this blog. Anyways, after tinkering and getting my demo server up and running, I decided to blog my experience.
The blog was primarily around my concept and theories on how MCP servers for observability were just hype machines. You can read the entire blog here.
About a day after it went live, we saw a sudden (and unexpected) surge in traffic. We hadn't even distributed it yet! On tracking down the source by breaking-down via custom-intial-referrer, we found that our blog was featured on the TL;DR newsletter.
Now, I'm a very pessimistic person and I believe nothing good ever happens to me. So I was naturally inclined to believe that a bot which intermittently scrapes the web, found my article based on some keyword trigger and pushed it to the site.
Not being able to find another reasonable explanation, I went with that story.
Cut to, two days later, I woke up and was on my bed scrolling through every social media site. My career choices have definitely made LinkedIn one of the first sites I srcoll across. Very surprisingly, I found a DM from Andrew Tan.
It's not everyday that you wake upto a DM from an investor at both OpenAI and Anthropic and the founder of TL;DR AI himself.
And I will forever hold this DM close to my heart. Here's a screenshot.
I've been coding, building, and writing for a while now. In moments like these, all those times I questioned my choices finally fall into place.
Full circle moment.
Long story short, an article that I authored, went viral on Hacker News (HN). When I started, I hardly had any idea of what HN was and what it meant to be on it's front page. I realised it was insanely hard to make it up there, when I started attempting to post some articles there.
For those who understand math over text, here are a few stats as brainstormed by Grok.
Estimates suggest there are about 1,300 submissions per day, and only a small fraction achieve significant attention, such as over 100 points, which is considered viral. This leads to an estimated probability of 1 in 1,000, though this is a rough figure and can vary.
To elaborate on the benefits we got from making it to the front page, apart from feeling great, take a
look at the mixpanel dashboard below.
The highest peak seen in the image, is from the day the article made it to the frontpage. We got a single-day traffic of 5K users and it continued to grow. We created a momentum which kept propagating for many more weeks! Today the article has about 15.8K pageviews.
And that is the power of HN.
The article was titled CI/CD Observability with OpenTelemetry. Yep, not even a click-baity, catchy title. You can read the entire article here. It explains how you can configure your OpenTelemetry collector to collect traces from your Github Actions CI/CD pipeline.
I'll also attach a screenshot of the article featuring on the frontpage of HN.
This gained traction and we ended up collecting 150+ upvotes and 50+ comments.
Now if you ask me for tips to do the same, I don't have any. There's no playbook for HN. It's pointless to spent time figuring one out as well. You can have playbooks for almost every other platform, but surely not for HN, which is exactly what makes it nearly impossible to make it to teh frontpage as well.
The only tip/ general gyan I can give you is to put your 200% into what you build and write.
Try your best, for the rest will fall into place.