top | item 44855157

Show HN: Engineering.fyi – Search across tech engineering blogs in one place

470 points| indiehackerman | 7 months ago |engineering.fyi | reply

I built a search engine for engineering blogs because I was tired of manually checking individual company blogs to find real-world production examples.

The problem: When learning a new technology, the best insights often come from how companies like Google, Meta, or Stripe actually implement it in production. But these gems are scattered across dozens of separate engineering blogs with no way to search across them.

What I built: Engineering.fyi indexes engineering blogs from ~15 companies (Google, Meta, OpenAI, Anthropic, Stripe, Uber, etc.) and makes them searchable in one place. You can filter by topic, difficulty level, and whether articles include code samples.

Technical details: - Built with Next.js, SQLite, DrizzleORM - Custom scrapers for each blog (they're all frustratingly different) - Basic tagging system using content matching (still improving this)

Current status: Core search is working. Adding new blogs weekly as I index them.

Next features (based on early feedback): - AI summaries for quick article previews - Weekly digest of trending engineering insights - Save/bookmark articles (considering whether to add accounts)

Interesting challenges: - Each blog requires custom parsing logic (no standard format) - Building an accurate tagging system is harder than expected – started with exact matching but exploring better approaches

I'd love feedback on: - Which company engineering blogs you'd find most valuable to include - Whether AI summaries would actually be useful or just noise - How you currently discover engineering articles from these companies

126 comments

order
[+] freetonik|7 months ago|reply
Shameless plug, but hopefully relevant enough: my directory and search engine for personal blogs[1] indexes over 1000 RSS feeds, and naturally lots of them are about engineering and software development. Full-text search is implemented with Typesense, and there are also "related" recommendations for each post, example [2].

1. https://minifeed.net/

2. https://minifeed.net/items/n1HZYMDEKyra

[+] pnt12|7 months ago|reply
I was surprised that I started reading one blog on your site, and it was badly written. Only afterwards below the text, did I notice it was a summary!

I'd advise to put it at the top, before the text, to let people know beforehand and not be caught off guard. Then you can have a big button saying "read full article in website" or something, to make it easy for people to see both options.

[+] jpmonette|7 months ago|reply
Looks really nice! Any plan to add social aspect like comments, likes and such?
[+] gombosg|7 months ago|reply
I kind of miss the RSS days when you just had your own news/blog aggregator without the annoyance of Substack, Medium or anything else.
[+] rambambram|7 months ago|reply
Don't act like RSS is done with. It's twenty plus years old, and still going strong. Nobody is stopping you from using it, whether you only read or also post.

Hyped up tech is like milk, it stinks after a couple of days. Open protocols are like fine wine, they age beautifully.

P.S. Your site is offline. If it wasn't and you even had one interesting article, I would have added your website to my list of feeds. I picked up hundreds of interesting websites/feeds through HN alone in the last years.

[+] dewey|7 months ago|reply
Less people are blogging these days but there's still a lot of interesting blogs out there. It's even more self-selected than before but I almost always find a RSS feed for a blog that I think is useful and interesting.
[+] pbronez|7 months ago|reply
Cool idea. I thought I’d try to make a Kagi Lens to accomplish the same thing:

https://kagi.com/lenses/LdYine8hZtYmrt8yTMngOUtvTM9rmkRy

Kagi Lenses can be defined in many ways, one of which is specifying URLs to search. Unfortunately you can only provide 10 URLs per lens. Here are the ones I chose:

https://stripe.com/blog/engineering, https://engineering.fb.com/, https://www.uber.com/en-US/blog/engineering/, https://netflixtechblog.com/, https://research.google/blog/, https://technology.riotgames.com/, https://incident.io/blog, https://www.anthropic.com/engineering, https://openai.com/news/, https://shopify.engineering/

[+] pbronez|7 months ago|reply
When I use this lens to search for “Python” the top three hits are:

Meta’s Pyrefly announcement (may 2025)

Netflix post about their overall use of python (March 2013)

Google’s announcement of the Croissant ML metadata format (March 2024)

[+] __turbobrew__|7 months ago|reply
No AWS blog? If anything I have found the AWS blog the highest quality and most novel. The articles on things like route53 are really interesting.
[+] jzig|7 months ago|reply
Now make a lens of lenses!
[+] kenanfyi|7 months ago|reply
Looks good, but it‘s fascinating the term engineering nowadays almost only boils down to software(also mostly web) and AI, although it is way more than that.
[+] mrugge|7 months ago|reply
the engineering that pays (big bucks)
[+] mustaphah|7 months ago|reply
It would be awesome to have a newsletter highlighting the top X articles (fully automated). You could start with a simple scoring system (page views) and maybe later add an upvote button so the most-voted articles get sent out each week.

Sending emails isn't cost-free, but AFAIK, Buttondown [1] has a free plan for up to 100 subscribers. It's dead simple: they provide an issue archive [2] and handle subscription for you [3].

With their 100-subscriber free plan, you could limit this feature to a close circle. Maybe later monetize the newsletter feature to cover the ESP costs.

[1] https://buttondown.com

[2] https://buttondown.com/hacker-newsletter/archive

[3] https://buttondown.com/hacker-newsletter

[+] demarq|7 months ago|reply
> When learning a new technology, the best insights often come from how companies like Google, Meta, or Stripe actually implement it in production.

I think not.

[+] ozgrakkurt|7 months ago|reply
Amazing idea but Cloudflare DDOSes my browser when I try to open it.

Also a nice reminder to move my website off of cloudflare asap

[+] jagged-chisel|7 months ago|reply
Maybe I’m a bit out of the loop on Cloudflare’s activity these days, but I’m not sure I understand your statement.

Cloudflare thinks your browser is part of a DDoS?

Cloudflare is attacking your browser from several places across the internet?

[+] skhameneh|7 months ago|reply
Can you add cross support for Fediverse? I really haven't been keeping up but I think with ActivityPub you can support Bluesky, Mastodon, and Threads. A lot of the general support and progress went from GNU Social > Mastodon > whatever.
[+] idosh|7 months ago|reply
Shameless plug, you can find all of them and many more on https://daily.dev/. It's a personalized aggregator for developer news
[+] grub5000|7 months ago|reply
I can't figure this one out - is it only a browser extension? The site keeps trying to trick me into installing a browser extension, which seems incredibly sketchy
[+] mmargenot|7 months ago|reply
Are you trying to stick with company blogs primarily, or to expand into general non-affiliated eng blogs? People like Maggie Appleton (https://maggieappleton.com/) and Patrick McKenzie (https://www.kalzumeus.com/) frequently have compelling ideas around technology, but I suppose that's a different "product" from what a company blog is selling.
[+] indiehackerman|7 months ago|reply
Great suggestions, I'm a big fan of Patrick McKenzie as well. I was going to start with company blogs but expand to individuals as well if there's enough demand/usage
[+] chandu_vrs|7 months ago|reply
Nice work. I started working on something similar, but my use case is slightly different (but the solution is similar). We all are interested in certain topics, there is a lot of content that gets published and our time is limited, so we need something that helps us identify top 10 articles or so per topic. This is why most of us like HackerNews. I think HackerNews by topic or interest would be a good idea to implement (but along with users posting links, it can come from crawling few sites as well)
[+] 8organicbits|7 months ago|reply
> Each blog requires custom parsing logic (no standard format)

This is unfortunate, RSS has promise to be that standard format. I've seen high adoption, but it's not universal.

[+] indiehackerman|7 months ago|reply
I started off with scraping through RSS but quickly realised it doesn't include all historical posts
[+] Catbert59|7 months ago|reply
First thing I saw was an annoying Cloudflare captcha.. ugh
[+] fuzball1989|7 months ago|reply
Are you sure you want to add hundreds of blogs? I would keep it curated to 10-20 otherwise it will turn into an RSS feed but I think you are chasing a goal of having the most interesting blogs to read and for people to use in their designs, coding etc.
[+] indiehackerman|7 months ago|reply
I think people would get less value if I kept it to 10-20. I was thinking of extending it to have user accounts were people could create their own lists of the articles (based on company, author, tags, etc) on there if there was enough interest