Hello, alterego
We built alterego because the existing tools for "online discovery" force you to pick a side: listen (Mention, Brandwatch, Octolens) or publish (Buffer, Hootsuite). Nobody connects them with judgment in the middle.
The gap
A founder watching Reddit for posts about her product faces three problems at once:
- Surfacing the conversation — keyword matching produces hundreds of hits a day, most of them noise.
- Deciding when to show up — most posts aren't actually about her product; some are, but with the wrong intent (rage-bait, sponsored shilling); a small minority are real opportunities.
- Drafting what to say — replies that lead with "we built X to solve exactly this" get downvoted within minutes; replies that lead with genuine value get upvoted and remembered.
These are three skills. The first is data filtering. The second is judgment. The third is writing. Each has different failure modes.
What alterego does
Three stages, each with a clear job:
Discovery monitors Reddit, Hacker News, and Bluesky for pain-point language matching your brief. The brief is a small product description plus keywords, watched subreddits, and a "do not engage" list (competitors, ban-prone topics). Polling runs every 15 minutes for HN and 20 for the other platforms.
Diagnosis scores each candidate post on four axes: problem relevance, purchase intent, urgency, and plug-risk. The fourth is the load-bearing one — it's why the same post can be a real opportunity for one product and a moderator-flag for another.
Drafting generates a reply that leads with value, discloses affiliation explicitly, and matches the platform's register. Every draft gets a risk panel (banned-claim matches, compliance score, recommendation) that a human sees before approving.
Compliance is the differentiator
The audit trail shows what was found, why it was flagged, what the model recommended, and what the human decided. Three reasons that matters:
- Reddit and Hacker News moderators will eventually ban anyone who looks like a spammer. The audit trail proves you didn't.
- Founders are the worst judges of plug-risk on their own product. A model that's calibrated against the social platform's norms catches the bad drafts.
- When the model gets it wrong (it will), you have a record of what it said vs. what you did. That's how you learn.
What's next
We're shipping the early Starter and Pro tiers ($49 and $199/mo) with a 14-day free trial — no credit card required during trial. We'd love your feedback on what's missing.