All activity
Gianmarco Carrierileft a comment
This acquisition makes total sense when you look at where the AI industry bottleneck is shifting. A year ago the conversation was "which model should I use?" Now it's "how do I serve it fast and cheap enough to build a real business?" I run multi-agent AI pipelines for travel planning (Aitinery) and inference cost is the single biggest factor determining what features we can actually ship. The...
Inferless is joining Baseten
fmerianJoin the discussion
Gianmarco Carrierileft a comment
This thread is incredibly timely — I'm preparing to launch Aitinery (AI travel planner for Italy) on PH in the coming weeks, and I'm learning some uncomfortable truths about community vs. audience. I have zero PH followers. Started from scratch a few weeks ago. So the "warm up your audience" advice doesn't apply to me. But what I've discovered is that QUALITY of engagement matters exponentially...
Can a large Product Hunt community help you with a product launch?
Gianmarco Carrierileft a comment
Here's the question nobody's asking: should you trust AI agents MORE than the humans currently handling the same tasks? Think about it. Your travel agent has access to your passport details, budget, family situation, and travel dates. Your accountant knows everything about your finances. Your doctor knows your entire medical history. We've been trusting HUMANS with sensitive data forever — and...
How much do you trust AI agents?
Gianmarco Carrierileft a comment
Honest answer? I push AI-written code to production every week. And I sleep fine. But here's the nuance nobody in this thread is addressing: the question isn't "do you trust AI to write code?" — it's "do you trust your ARCHITECTURE to survive AI's mistakes?" I'm building Aitinery (AI travel planner, Django + Next.js) and here's what I've learned: AI-generated frontend code is almost always...
Do you trust AI to push code to production?
Mason BachmannJoin the discussion
Gianmarco Carrierileft a comment
This is the most important post I've read on PH forums this month, and I think you're actually UNDERESTIMATING how radical the shift is. Here's what nobody's talking about: we're not just redesigning interfaces for agents — we're building a world where agents talk to agents, and humans are just... spectators with veto power. I'm building AI travel agents (Aitinery — multi-agent pipeline using...
It’s time to start designing for agents, not humans
Max MusingJoin the discussion
Gianmarco Carrierileft a comment
Your responsibility question is THE question that nobody in AI wants to answer honestly. Here's my unpopular take: the liability framework we're trying to apply to AI medicine is fundamentally broken because it assumes AI works like a human doctor — it doesn't. When a doctor misdiagnoses, we understand the failure mode: fatigue, cognitive bias, limited knowledge. We have centuries of...
Use of AI in medicine – 3 projects that show it's already happening
Gianmarco Carrierileft a comment
Hot take: the "N8N + Lovable baby" comparison actually undersells what's interesting here. The real question nobody's asking about no-code agent builders: what happens when your agent needs deep domain expertise? I've been building AI travel agents (Aitinery — we use Google ADK with multi-agent pipelines for Italy trip planning), and here's what I learned the hard way: the orchestration layer...
Architect by LyzrBuild AI that works for you
Gianmarco Carrierileft a comment
@peter_albert nailed it. I'm running Gemini models in production for Aitinery (AI travel planner) and this is exactly the gap. Benchmarks say Gemini is world-class. My production logs say it sometimes hallucinates restaurant names that don't exist and occasionally generates itineraries with 16-hour driving days. Benchmarks don't test "can this model reliably plan a family trip to Puglia without...
Gemini 3.1 ProA smarter model for your most complex tasks
Gianmarco Carrierileft a comment
This is fascinating research, but I want to push back on something everyone seems to be celebrating uncritically. Are we sure we WANT AI assistants to become the primary product discovery layer? Because that's what AEO is really building toward — and as a maker, this terrifies me more than it excites me. Here's the uncomfortable question: if ChatGPT becomes how people discover products, then...
Case Study: how Product Hunt can improve AI visibility in 2026
Andrew StewartJoin the discussion
Gianmarco Carrierileft a comment
Hot take: being nervous about your PH launch is actually the problem, not the solution. Here's what nobody in this thread is saying — the marketer with 10k followers who finished 6th? They probably treated PH as a marketing campaign. That's the trap. Product Hunt isn't a campaign. It's a single data point in a 10-year journey. I'm building Aitinery (AI travel planning) and I'm also preparing to...
Launching on Product Hunt Next Week... and Honestly, I'm Nervous
Imtiyaz Join the discussion
Gianmarco Carrierileft a comment
Love the "one prompt" approach to lead gen — it's the same philosophy we follow at Aitinery for travel planning. Describe what you want, and the AI does the heavy lifting across multiple data sources. The 100+ data sources integration is impressive. Curious about the enrichment accuracy — how do you handle conflicting data across different sources? That's always the tricky part when aggregating...
Origami.chatFind your perfect leads with one prompt
Gianmarco Carrierileft a comment
Really interesting data on the CS enrollment shift. As someone who studied computer science and now builds AI products, I see both sides of this. The shift from "learning to code" to "learning to think with AI" is real, and I think it's actually healthy. When I'm building Aitinery (AI travel planner), I spend way more time on system design, prompt engineering, and understanding user problems...
What will the future of studying be like when AI does everything for us?
Gianmarco Carrierileft a comment
This hits close to home. I'm building Aitinery (AI travel planner) on Next.js + Tailwind CSS 4, and I realized I almost never visit the Tailwind docs anymore — Cursor and Claude generate the utility classes directly. I'm literally part of the problem this thread describes. The paradox is brutal: the better AI gets at generating framework-specific code, the less we interact with the original...
Vibe coding is thriving. The tools powering it are quietly dying.
Tanzil ChowdhuryJoin the discussion
Gianmarco Carrierileft a comment
Great follow-up to the trust thread! As someone who both uses AI daily and builds with it (I'm working on Aitinery, an AI travel planner), I think about this from both sides. As a user: I share way more than I probably should. Code, business ideas, personal frustrations, brainstorming sessions. ChatGPT has basically become my thinking partner. The convenience always wins over the privacy...
How much do you share with AI? Or more specifically, with ChatGPT?
Gianmarco Carrierileft a comment
Here's mine — curious to see if you can guess it! "Your AI travel team for Italy — from hidden gems to real train times." (Bonus hint: it's not a chatbot, it's a team of specialized AI agents that collaborate)
🔥 Drop your tagline and I'll try to guess what your product is
Aaron O'LearyJoin the discussion
Gianmarco Carrierileft a comment
This is incredibly valuable, thanks for sharing your whole plan! I'm preparing for my own PH launch (Aitinery, an AI travel planner) and the point about "relying on follower count alone is not enough" really hit home. I've been focused on building genuine connections in the community rather than just chasing numbers — commenting on relevant products, participating in discussions, and actually...
Can a large Product Hunt community help you with a product launch?
Gianmarco Carrierileft a comment
Love this format! Here are my Q1 2026 goals: Work goals: Launch Aitinery (my AI travel planner for Italy) on Product Hunt — this is the big one! - Reach 500 organic users/month through SEO (just published 20 blog articles, fingers crossed) - Ship the "Italy Specialized AI Engine" — a major upgrade to our multi-agent pipeline with real train schedules, regional cuisine data, and local hidden...
2026 and your goals. Let's try to set them for Q1.
Gianmarco Carrieristarted a discussion
Hey PH! I'm Gianmarco, a software engineer from Italy building Aitinery
Hey everyone! I'm Gianmarco, a software engineer based in Italy. After years of planning trips for friends and family (and being the unofficial "travel agent" of my group), I decided to build something that could do it better than me — or at least faster. I'm working on Aitinery (aitinery.com), an AI-powered travel planner that specializes in Italy. The idea is simple: instead of spending hours...
Gianmarco Carrierileft a comment
As someone building with AI agents every day (I run Aitinery, an AI travel planner), I'd say trust depends entirely on the domain and the guardrails you put in place. For low-stakes, creative tasks like planning a trip itinerary, I trust agents quite a lot — if the AI suggests a slightly wrong restaurant, no real harm done, and you can always adjust. For finances or health? Hard pass, at least...
How much do you trust AI agents?
Gianmarco Carrierileft a comment
The map integration is a great differentiator! As someone also building in the AI travel space (Aitinery, focused on Italy), I've found that showing places on a map changes everything for the user experience. People think spatially when planning trips, and a plain list of POIs just doesn't cut it. The conversational UI approach is interesting too — how do you handle it when users ask for very...
Copilot2tripPersonalized AI-powered travel assistant with maps