Ophelia

Intent will swallow the entire stack of execution. Swallow agent merchant access, swallow cards, swallow all of it. He is right.

you can now text an agent "purchase the items in my @amazon cart now, and add a pair of @UGG boots" and it just…completes the request! proud to bring @opheliaapi to every agent on @onkeiki. restaurants, flights, hotels, tickets, retail, grocery shopping. takes a few minutes to set up. builders, go cook. intent is the new interface.🦾😤

We're proud to be working with the @orchid_hq team to bring @opheliaapi into Keiki’s platform. Now, anyone's text-line agent can complete checkout across restaurants, flights, hotels, tickets, retail, and 1700+ merchants, all with confirmation directly in the thread. They’re one of the only teams thoughtfully building the easiest way to launch an agent people actually want. And now it’s even easier to transform intent into an actual outcome. Let it rip and send me what you book. onkeiki.com/signup

locked in until IPO or exit. until then AI will save us. Betaworks looking pretty tough with @OpheliaAPI merch 😤

Browser use is not the solution. We benchmarked @OpheliaAPI vs Pomerado vs @OpenAI Codex on the same workloads, same sites, same queries, same starting conditions. Our goal: measure latency, inventory depth, price accuracy, and transaction completion. 1 / OpenTable: latency + inventory parity - Ophelia: 12–20s - Pomerado/Codex: 43–45s - Every bookable restaurant returned the same available time slots across runs → ~2–3× lower median latency 2 / Hotels: retrieval depth + price accuracy - Ophelia: 96–100 results - Pomerado/Codex: 20 Ophelia traversed beyond the first page and surfaced lower price floors in NYC and Chicago. Their displayed hotel prices also excluded tax. 3 / Flights + Resy: coverage + latency They also missed a $120 BOS → MIA nonstop that Ophelia surfaced. The important distinction: Browser agents optimize for navigating a website. Truly just "search". But Ophelia optimizes for discovery, retrieving inventory, and actually completing the transaction. The benchmark should never be only speed, that's inconsistent and unreliable. Rather who can actually complete purchases safely, reliably, and at scale.

A thing we’ve been working on for several years: restaurants, flights, hotels, tickets, and retail booked from your product, merchant confirmation (and receipt sent directly to inbox!), on @opheliaapi. Show up for the VC coffee; table's already under your name. Power move - not a link, redirect, or “here’s what you could book.” No more building wrappers in 2026 – at San Francisco, CA

this will all make sense soon. POV: the dungeon at @betaworks is basically a rite of passage at this point. V3 is an unreasonable amount of engineering. no one is ready for it.

on a serious note: our team takes real pride in the craft, we care what we build. And that their products actually work. love how a lot of these companies already book and actually facilitate dates through @opheliaapi. a lot of people assume “dates” are just reservations, except they’re also activities, concerts, gifts, the entire night.

numbers i’d want to see from api "routing layer" this week: to every new launch: show me the merchant confirmation, show me checkout, stop calling catalogue search wrappers “infrastructure” it’s misleading and fraudulent at the least. sick of the lack of creativity and insane copying from certain teams.

the perfect use case for founders + engineers who are looking to solve human connection with social and lifestyle apps. consumer is the only way to solve everyday living and life. you have the power - the tool is in your hand. dating apps with reservations, friend hangouts with tickets, workout classes for group meetups. it’s all there. you have the api. go build. create the world you want. opheliaos.com

life is only fun when you get to use tech to build things that get you off tech. bless @opheliaapi

nyc dinner reservation made entirely through an audio-only hardware interface. glimpse at the future AGI-powered purchasing. credit: juan from @amex built with @opheliaapi

We’re so excited to announce we’ll be joining the @SPECS team in Brussels this December to hack with their AR + AI team! Every computer so far has needed you to stop and look down at it, type, or manually click. We're building @OpheliaAPI so that the mechanic who’s hands are dirty can order parts before forgetting. For when you’re walking past a restaurant and saying "book a reservation here tomorrow at 8." Or glancing at a concert poster and the tickets are already yours. No tabs, no manual checkout. You see it, you say it, it’s done. That's the AGI we're pushing for: everyone can experience, and every builder has fun. The power is yours. See you soon Brussels :)

respectfully, i disagree that an ai has to choose between caring about you and doing things for you. the mom at 3am with a newborn wants to hear she's doing great. she also wants size 1 diapers at her door by 7am, and the formula her pediatrician named, in stock, in the right size. a warm reply is one inference call, the cheapest thing a model produces. the hard part is everything after it: live inventory, the exact variant, a confirmation that comes from the merchant, and a retry that doesn't double the order when the wifi drops at 3am. the metrics here are time spent and messages sent. those measure attention. progress toward agi shows up as things that got done for people. caregivers are drowning in logistics. support that can't place the order stops at sympathy. the most caring thing an ai companion can do at 3am is take something off the list.