Resolve bank customer chats with an AI chatbot
Commerzbank's Bene chatbot, now on Gemini, handles over two million chats and resolves 70% of inquiries.
Original post ↗Real workflows with the tools, steps and prompts to copy them. Filter by what you use and how much time you have.
Commerzbank's Bene chatbot, now on Gemini, handles over two million chats and resolves 70% of inquiries.
Original post ↗A creator tested Gemini Omni Flash over 36 hours, editing clips in place from typed instructions while preserving lighting and camera angle, with variable quality.
Original post ↗The Gemini Plays Pokémon agent reads game RAM and screenshots, turns actions into button presses, and uses pathfinder and boulder-puzzle sub-agents.
Original post ↗A developer wired Gemini's Computer Use model to Playwright in Java and had it play a simple button-clicking game to verify click-coordinate scaling.
Original post ↗A reporter linked OpenTable, Ticketmaster, Uber and Thumbtack to Alexa+ for booking, but found it beta-like, error-prone and sometimes not hearing them.
Original post ↗Google's Big Sleep agent reviewed recent SQLite commits starting from a patched bug and found an exploitable flaw, which was fixed before release.
Original post ↗Gemini detects objects and writes robot API code, and with 10 in-context demos an ALOHA robot packed boxes and folded a dress; voice commands became robot plans.
Original post ↗Gemini Robotics-ER 1.5 looks up local recycling rules online, then plans and sorts objects into compost, recycling and trash bins.
Original post ↗Gemini Robotics ER 2 drives a Boston Dynamics Spot, watching video to confirm tasks like tying a trash bag are done before moving on.
Original post ↗Alexa+ summarizes notes sent from a Kindle Scribe and converts them into to-do lists, calendar events and reminders; a feature description.
Original post ↗A reviewer tested Dots on product research, publishing a webpage, content planning and scheduled jobs; inbox triage and follow-ups needed supervision and a spreadsheet had broken formulas.
Original post ↗The author has a Dot named Jamie run a defined outreach campaign, keep the tracker accurate and flag replies or placements needing attention.
Original post ↗In OpenAI's demo, a Dot found a refundable policy-compliant flight and hotel options and prepared a booking, though the flight was still pending at the end.
Original post ↗A journalist tested Dot on ISP signup, an Ikea account and food ordering; it finished only some with human help, blocked by browser checks and payment steps.
Original post ↗A reviewer's dot flagged a failed payment unprompted but only read and reported it; a shopping site blocked it and follow-through was unfinished.
Original post ↗An educator's Dot checks public documents for changes, sorts them by impact, and verifies an article's notation against linked pages, offering revision candidates only.
Original post ↗A user had a Dot monitor rare US coffee beans twice a day using context from old chats, and it returned Gesha options priced per cup.
Original post ↗The author had a Dot find footage, build the B-roll edit, add captions and arrange visuals over several review passes, working from his recorded narration.
Original post ↗