All activity
Alexandre Poitevinleft a comment
Zero LinkedIn bans. It's not a marketing slogan — it's a technical obsession. We spent months calibrating delays, randomizing actions, respecting rate limits. Your account security is our #1 priority.
XP OneReplace Waalaxy, Lemlist, LGM, Apollo, Hunter & PB → $29/mo
You’re probably paying for Waalaxy, Lemlist, LGM, Apollo, Hunter and PhantomBuster.
That’s $377+ per month for 6 tools, 6 logins and 6 invoices, and they don’t even work together.
XP One replaces all of them.
One Chrome extension. One dashboard.
Collect leads from LinkedIn in one click.
Enrich them with verified emails and phone numbers.
Launch campaigns on LinkedIn, Email and WhatsApp.
Track everything in a built-in AI CRM.
One tool. One flow. $29 per month.
More clients.
XP OneReplace Waalaxy, Lemlist, LGM, Apollo, Hunter & PB → $29/mo
Alexandre Poitevinleft a comment
The claim about keeping characters consistent across a full video is honestly the most interesting part because that’s where most prompt-to-video tools fall short. But I couldn’t find any real detail on how it works technically — is it reference images stitched across frames or something happening at the model level?
LunairGenerate studio-quality explainer videos. Instantly.
Alexandre Poitevinleft a comment
They mention running open source HLE benchmarks to keep routing honest but theres no detail anywhere on methodology or cadence or how those results actually influence routing decisions per workload
ZenMuxAn enterprise-grade LLM gateway with automatic compensation
Alexandre Poitevinleft a comment
You mention HIPAA compliance and encryption at rest, but I didn’t see anything about a BAA, SOC2, or real clinical validation yet since it’s still “launching soon.” What has the crisis detection system actually been tested against so far?
Lovon AI TherapyTalk it out and feel better
Alexandre Poitevinleft a comment
Matching against 1.3B+ profiles, is that actual vector similarity on enriched embeddings in Pinecone or more rule-based filtering with AI reranking on top? Big difference for niche ICPs.
StarnusFind and reach your next customers on autopilot
Alexandre Poitevinleft a comment
Voice cloning that preserves the original speaker's identity in real time across 150+ languages is no joke technically. The dialect-level support (LatAm vs Castilian Spanish, Canadian vs Parisian French) is a nice touch too. Are you doing the speech-to-speech translation end to end on MARS8 or is there still a separate STT step in the pipeline?
DubStream by CAMB.AIDub live streams in 150+ languages, instantly
Alexandre Poitevinleft a comment
The Engage Discover feature is interesting, finding trending posts to reply to before they go viral is a smart way to get organic reach. Curious how fast the trend detection updates.
SuperXAll-in-one growth OS for serious 𝕏 creators
Alexandre Poitevinleft a comment
Curious about the "private memory" claim for Clawdbot. Is it a local vector store, SQLite, something custom? Repository pattern?
Tinkerer ClubThe private club for ppl who automate, self-host, and use AI
Alexandre Poitevinleft a comment
The maker mentioned moving from regenerating charts per turn to "live charts where data can be refreshed." thats a fundamentally different architecture problem, stateless generation vs stateful component lifecycle. How is C1 handling component identity and diffing between turns today?
Agent Builder by ThesysBuild AI agents that respond with UI instead of text