For about two years my "keeping up with tech" system was a folder of 14 newsletters, three podcast apps' worth of queued episodes, and a YouTube Watch Later playlist that crossed 600 videos. I felt informed. Then a coworker asked me what had changed in React 19 and I couldn't name a single thing, even though I'd read at least four posts about it.
That's when I figured out the problem. I wasn't short on information. I was consuming it in a way that left nothing behind.
If you've searched "how to keep up with technology as a developer" and found lists of 50 newsletters to subscribe to, this is the opposite advice. Fewer inputs, and a way to keep what you take in.
You don't need to know everything that shipped
Nobody keeps up with all of it. Between framework releases, new databases, AI model launches every other week and whatever Hacker News is angry about today, the volume is beyond what one person can read, let alone use.
What actually helps your career is narrower. You want to know the changes that touch the stack you work in, a rough sense of what's emerging nearby, and deep knowledge of a few things you care about. Everything else is trivia you can look up the day you need it.
So the first step is deciding what you're keeping up for. I wrote mine down: TypeScript and React because that's my job, Postgres because I keep getting pulled into query tuning, and LLM tooling because I build with it. Anything outside those three got a much harder filter.
Cut the inputs down hard
I unsubscribed from 10 of the 14 newsletters. The ones I kept were the ones I'd actually opened most weeks: a weekly JavaScript roundup, one Postgres-focused newsletter, and two written by individual engineers whose taste I trust.
Same with podcasts. I went from following around 20 shows to four. If I'd skipped the last five episodes of something, it was already telling me it wasn't important.
A quick test that worked for me: for each source, ask "when did this last change how I wrote code or made a decision?" If you can't remember, it's entertainment. That's fine, but file it as entertainment and stop feeling guilty about the backlog.
Batch it instead of grazing
Grazing is reading a tech article between Slack messages, half a podcast on the commute, a YouTube talk at 1.5x while eating. It feels productive and it's the main reason none of it sticks. Your attention is split, and nothing connects to anything else.
I now do one 45-minute block on Friday afternoon. I go through what I saved that week, read or watch the few things that still look worth it, and delete the rest without opening them. Most weeks I delete more than half. Things that seemed urgent on Monday rarely are by Friday.
Podcasts still happen on walks, but only from the short list.
Write one sentence per thing you finish
This is the habit that made the biggest difference, and it's tiny. After every article, talk or episode I finish, I write one sentence about what I'd tell a teammate. Not a summary. Just the one thing.
React 19: Actions let forms handle pending and error state without useState boilerplate, and
use()can read a promise inside render.
If I can't write that sentence, I didn't really learn anything, and I either go back for the part I missed or let it go.
These sentences go into a single running file. After a few months it's the most useful document I own, because it's my own words about stuff I actually cared about.
Come back to it, or it's gone
Here's the uncomfortable part. Even with a good note, most of what you read fades within weeks unless you see it again. That's the forgetting curve, and it's been measured since Hermann Ebbinghaus ran memory experiments on himself in the 1880s.
The fix is spaced repetition: review something a day after you learn it, then a few days later, then a week, then longer gaps as it sticks. I wrote about implementing the SM-2 spaced repetition algorithm in JavaScript if you want to see how the scheduling works.
For tech news, I don't need flashcards on everything. I need the handful of things that matter to resurface at the right times, so that when a coworker asks about React 19, I've seen my own note about it three times already.
Where Recl fits in
The weak spot in my system was always the saving step. Videos went to Watch Later, articles to a read-later app, podcasts stayed in the podcast app, and Instagram reels from dev creators went into a saved folder I never opened. Four piles, none of them talking to each other, and none of them reminding me of anything.
That's the gap I built Recl for. You share a video, reel, podcast or article to it from your phone or the Chrome extension, and it writes a summary of the key ideas (with timestamps for videos), sorts it into a topic, and brings it back 1, 3, 7, 14, 30 and 60 days after you saved it.
The feature I use most for this is Ask. I can type "what did I save about Postgres indexing?" and get an answer drawn from my own saves, with the sources, instead of digging through browser history for the talk with the good EXPLAIN ANALYZE example.
It doesn't replace the one-sentence habit. I still write those. Recl just makes sure the stuff I saved doesn't vanish the moment I close the tab.
What a normal week looks like now
Monday to Thursday, I save things and mostly don't read them. Friday afternoon I spend 45 minutes on the pile, finish a few, write a sentence for each, and delete the rest. Reviews show up on their own schedule and take a couple of minutes a day.
I read less tech content than I did two years ago, probably by half. I also remember a lot more of it, and I can finally answer the "what changed in X" question without bluffing.
I build Recl, an AI learning app that saves your videos, reels, podcasts and articles, summarizes them, sorts them into topics, and uses spaced repetition so you remember what you learned.
Top comments (0)