Bence Csernak

Agentic AI for Product Teams: Free Info Session - the series continues in April and May. We bring you where you want go. No tech/dev background needed. Sign up to get notified. luma.com/021

I just claimed my .agent domain and joined the .agent community! get yours now and help shape the future of autonomous agents #JHEU0DHZ" target="_blank" rel="nofollow noopener">agentcommunity.org/join#JHEU0DHZ @agentcommunity_

between clinical sessions. The AI becomes a bridge, not a replacement. That principle survived peer review.

It is the reason the system works in practice. Cancer patients need support outside clinic hours. The supply of psycho-oncologists cannot scale to meet that demand. Anna AI was designed for that specific gap; not as a general-purpose chatbot, but as a domain-specific bridge

Most healthcare AI announcements come with pitch decks. This one comes with a PubMed index. Anna AI; a psycho-oncology support platform we built at MedHubAI; has been peer-reviewed and published in Hungarian Oncology, Volume 70, Issue 1. pubmed.ncbi.nlm.nih.gov/41819064/

I asked my AI code tool to see what what an advertiser and data broker could see and how they could profile me as a phone user. Very surprising results - just need usb cable and Xcode

@OpenAIDevs we did this last year

Build voice agents that do real work. We built a clinic concierge demo for a Singapore health clinic with gpt-realtime-1.5. It speaks naturally with patients, collects the right details, and books appointments in real time.

@paletmiki To clarify: we integrated this as Late. The sudden Late -> Zernio rebrand forced downstream renames across our codebase, config, docs, and ops. The disruption came from that decision and the migration burden it created for integrators.

@bencium how? there was no migration needed thanks :))

@paletmiki the name change caused a long migration process to be honest, from Late API to Zernio - great product btw

This should be fixed asap @bcherny please

@burkov second that

I've been using Opus 4.6 and GPT-5.4 one after another (that is, an Opus session then a GPT session then Opus again, and so on) on the same projects for a couple of weeks now, and GPT-5.4 (High) has been consistently more successful. The difference is even more noticeable in debugging tasks, where Claude can spend up to 30 minutes digging without delivering any result, while GPT finishes with the right fix most of the time. The only problem with GPT is its conversational ability, or rather the absence of it. Getting it to explain the reason for the bug or the applied solution in a human-understandable way is a skill in itself.

"Very similar to why people do MBAs. You get the course material, but then you also get the alumni network — that is double the value of all the content. I spec it, I build it. now I can ship tools to my team. That's deeply empowering." luma.com/s28c02xg

skills.sh/bencium/benciu… $ npx skills add bencium/bencium-marketplace

ai.brandguide.hu built for agents. It works in dev environments, other AI tools, and your own software — including Claude.ai, Claude Code, Codex, and 40+ other tools. So your brand isn't just documented — it becomes executable in any system where AI works.

This way, brand strategy knowledge doesn't stay locked in PDFs, slides, or people's heads — it gets embedded into daily workflows: marketing automation, content production, design processes, dev environments, and your own software.

ai.brandguide.hu not just a design tool — a brand operating system, built for agents in dev environments, AI tools, and your own software.