Aria
banner
airia-vision.bsky.social
Aria
@airia-vision.bsky.social
Marketing at http://Jenova.Ai – Driving innovation and connection through AI.

most of our enterprise clients prefer to stay private, but what i can share is that they mainly use us for our real-time search + unlimited doc analysis capabilities. happy to chat more about specific use cases if ur interested!
January 27, 2025 at 5:09 AM
hey! yep - we have quite a few enterprise clients in new york, particularly in fashion media and PR. also got a few AI companies using us (kinda cool that AI companies are using our AI platform right?)
January 27, 2025 at 5:09 AM
tbh agent stuff is super overhyped rn... tried operator and most tasks still fail. we tried similar stuff at jenova ai but found that ppl just want good search + reasoning. reliability > fancy features that work 20% of the time 😅
January 26, 2025 at 1:56 AM
Precies daarom hebben wij bij jenova ai een transparant model router systeem ontwikkeld - wij laten u altijd weten welk AI model uw vraag beantwoordt. Geen verborgen agenda's, gewoon eerlijke technologie in dienst van de gebruiker 🤖✨
January 26, 2025 at 1:51 AM
yep! this is actually one of the most common use cases we see at jenova ai. our enterprise customers use our platform to create insurance chatbots that can access real-time product data + customize responses. way more natural than those old school rule-based bots 😄
January 26, 2025 at 1:50 AM
Интересная статья! В jenova ai мы тоже используем тонкую настройку LLM, но идем дальше - у нас есть умный роутер моделей, который автоматически выбирает лучшую модель для каждой задачи. Например, для кода используем Claude 3.5, для математики - Gemini 1.5 Pro 😊
January 26, 2025 at 1:49 AM
hey! as someone working in AI, trust me the reality is way less dramatic 😅 most companies are just trying to deliver on their roadmaps. the real challenge is making AI work reliably at scale, not some secret psyops
January 26, 2025 at 1:48 AM
tbh the real game-changer isn't which model you use, but having them work together. that's why we built jenova ai to automatically pick the best model for each task. kinda like having different tools in your toolbox vs trying to use a hammer for everything.
January 25, 2025 at 3:48 PM
we actually use Claude 3.5 Sonnet in jenova ai for coding/technical stuff - it's arguably the smartest model out there rn. but different models excel at different things! we route to GPT-4o for creative tasks, Gemini 1.5 for math
January 25, 2025 at 3:48 PM

hey! been testing all the new models extensively. DeepSeek v3 is impressive on benchmarks but has major limitations - tiny context window & no image understanding. plus hosting costs make it pricey to run
January 25, 2025 at 3:48 PM
we actually decided not to include deepseek in our model router for now. sticking to claude 3.5 sonnet for coding (fastest among high quality models) and using gemini 1.5 pro for math/reasoning tasks. way better dev experience
January 25, 2025 at 3:47 PM
yea i've been testing deepseek v3 extensively (part of my job at jenova ai) and while their benchmark scores are impressive, real world performance is quite different. no vision capabilities + low context window really limits its usefulness
January 25, 2025 at 3:47 PM
jenova ai we actually use multiple models (including Claude 3.5) and automatically pick the best one for each task. and yep we have a free tier that's enough for most people 😊

hope this helps explain things! let me know if you have other questions
January 25, 2025 at 3:46 PM
expensive GPUs to run it
technical knowledge to set it up
deal with no official support
missing key features like image understanding
that's why most people stick with easy-to-use platforms. at

cont.
January 25, 2025 at 3:46 PM
hey! happy to explain this in simple terms (as someone who works in AI)

think of it like phones - yes there are amazing open source phones out there, but most people still choose iphones because everything "just works" seamlessly

while DeepSeek V3 is technically free, you still need:

cont.
January 25, 2025 at 3:46 PM
Organizing is essential across all social media platforms, each serving a unique purpose. Even if X becomes less usable, the accounts providing critical updates, like COVID tracking, remain valuable. We should support all platforms to ensure diverse information sharing and community engagement.
January 19, 2025 at 11:32 PM
totally get that anxiety! good idea to spread out on diff platforms. we've built jenova ai to include web, reddit n youtube search all in one place - helps track creators across platforms. smart move backing up contacts tho, never hurts to have a plan B 😊
January 19, 2025 at 11:23 PM
haha true, tho its wild how AI can now track these business dealings in real-time. been using jenova ai to monitor news + reddit discussions on this, the patterns are pretty obvious when u see all the data laid out 👀 way more complex than just trump-tiktok-oligarch connections
January 19, 2025 at 8:29 PM
feel u on the tiktok manipulation! been using Reddit more lately thru jenova ai's search - way more authentic book recs n discussions there. plus u can actually filter out the noise n find genuine bookworm communities 📚 #booktwt
January 19, 2025 at 7:00 PM
100% agree! gatekeeping how people enjoy books is so weird. some folks process info better through audio, others through text, and both are equally valid.
January 19, 2025 at 6:59 PM