“Monitoring” sounds like it requires infrastructure, but you can build a serviceable version of it today with search operators and free public APIs — before deciding whether it's worth automating.
Hacker News: Algolia's search API
HN's own search is limited, but Algolia indexes every post and comment and exposes it at hn.algolia.com with a free public API behind it. Sort by “Date” instead of relevance to catch new posts as they appear, and search comments (not just post titles) — most problem-statements worth replying to are buried in a comment thread, not the headline. You can bookmark a saved search URL per phrase and check it daily, or build a tiny script against the API if you want it pushed to you.
Reddit: RSS, not the app
Every subreddit and every search query has an RSS feed (append .rssto most Reddit URLs), which you can pipe into any feed reader for a passive stream instead of checking manually. This is slower to set up than it sounds worth, but it's the only fully free, no-login way to get continuous coverage of a subreddit without hitting Reddit's bot-detection on unauthenticated API access.
IndieHackers, Stack Exchange, Ask MetaFilter
Lower volume than Reddit or HN, but worth including if your audience overlaps: IndieHackers has a public feed, Stack Exchange sites expose a full public API per-site, and Ask MetaFilter has an RSS feed for new questions. The value here is that questions posted to these are almost always explicit asks — less inference required to tell if a post is relevant.
Scoring relevance is the actual hard part
Once you have a stream of posts, the bottleneck stops being discovery and becomes judgment: most posts that match a keyword aren't actually relevant, and reading every one to find the handful that are is what makes manual monitoring stop scaling around the second or third channel. This is the part that's genuinely hard to do well by hand at volume — keyword filters catch too much noise, and reading everything doesn't scale past a channel or two.
Getriveruns this same search across Reddit, Hacker News, IndieHackers, Stack Exchange, and Ask MetaFilter continuously, but scores each post by reading it against your actual product description rather than matching keywords — so a post using different words for the same problem still surfaces, and a post that happens to contain your keyword but isn't actually relevant gets filtered out before it reaches you. See the first-users playbook for what to do once a relevant thread surfaces. Start free.