I conceived and coded VeoVJ, a tool for live-remixing visuals generated with Veo, Google's generative video model. By setting the model's early 6-second outputs to music, I transformed them into cohesive audiovisual performances opening for major keynotes including Google Cloud Next, I/O Developer Keynote, and Google Marketing Live across 7 performances and three continents (seen by 6M+ online, 30,000+ live).
For Google I/O's mainstage, co-created and performed a live demo of Gemini Canvas features with Aaron Wade, building a functional vintage-style camera app in minutes that generated its own QR code on the fly.
In partnership with Google Developer Relations, pioneered the #30DaysOfGeminiCanvas challenge, live "vibe-coding" a new generative toy or tool every day for a month to demystify generative AI development. (Twitter)
Art-directed and generated the keynote countdown for Google Cloud Next using Veo, and produced the Google I/O 2025 opener with Creative Lab using seamless generation tech (7M+ viewers combined). (Cloud Next / Google I/O)
Shaped the voice and creative strategy for Gemini's social media presence across Instagram and X, developing novel prompting techniques and creative campaigns.
Developed experimental projects demonstrating creative applications of Gemini 2.0: Word-to-Code (p5.js sketch generator based on semantic word meanings) and Handspew (MediaPipe hand puppet interface giving AI commentary on surroundings).
Created the website doomers.fyi and developed Moral License for Matthew Gasda's play about Sam Altman, resulting in viral coverage in The Guardian, Airmail News, and The Free Press with sold-out runs in NYC and SF.
Developed 4 rapid prototypes in 5 days for a specialized client board exploring innovative applications of creative technology and generative AI.
Developed a demo of the Multimodal Live API for Gemini 2.0 Flash Experimental fetching real-time weather data and presenting it in user-selected creative styles. (Demo Video)
Featured panelist at ARTECHOUSE's inaugural session discussing the intersection of art, AI, social media, and emerging creative technologies. (Instagram)
Taught Introduction to JavaScript for Web Design in Simone Cutrí's Core Interaction class at Parsons School of Design. (Selected Slides)
Original visuals coordinator and creative director for FCUKERS, creating immersive live-coded visuals and AR installations for festival stages (co-directed with Maria Galebach).
Creative direction and real-time live-coded visual identities for multi-thousand attendee nightlife events in NYC.
Lead iOS developer for an augmented reality tour guide application utilizing geolocation technology, acquired by Syracuse University.
Led development of Jeffrey Alan Scudder's "Happy Hands" series: Wave (gesture mapping), Handtime (gesture communication), and Staka (gesture pinball).
Created interactive AR "magic mirrors" celebrating inclusivity and diverse body types for New York Fashion Week Fall and Spring 2023.
Developed a hands-free gesture fingerpainting landing page for IAMB literary magazine's issue "NOSTALGIA".
Artist-in-residence creating live-coded responsive visual experiences for musical performances at Pianos NYC.
Created low-poly AR portals for innovative 3D-printed footwear in collaboration with a Berlin/Dutch design team.
Created an AR try-on filter helping transgender individuals with body dysmorphia visualize specialized affirming clothing before purchasing.
Every day, Google's native image model Nano Banana imagines what I look like X days from an August 2025 selfie, aging me alongside the model through Instagram UI pastiche. A practice in natural language and self-surveillance.
A puritanical scorecard for human hedonism and AI promiscuity inspired by the Rice Purity Test. Surpassed 2 million users and viral reach.
One hundred video studies generated with Veo of AI-generated humans breathing, staging the uncanny encounter where minimal inhalation and exhalation unsettles our sense of the real.
A dating app that skips bios and prompts: screenshot your Instagram explore page to find your scrollmate via your digital micro-interactions and algorithm.
Uses multimodal reasoning models with native image generation to dream up what your husband (or wife, job, breakfast) looks like based on your selfie, reflecting training bias and synthetic perception.
Satirizing AI dating concierges that date for you: what happens when we optimize away the human elements of dating and every match is a total fumble.
An email lottery with a 1 in 1,000,000 chance that your message gets delivered, exploring digital communication, chance, and catharsis (5M+ views, 10,000+ entries).
A menu of online manifestation videos, frequencies, and guided affirmations allowing you to binge and listen to the cacophony of your every desire all at once.
Soft-sculpture installation for @pixelmouthnyc switching dynamically between global live streams (remote sea cameras, zoo animals, ASMR, Gaza bomb watches, UN meetings, and gallery security cams).
Curated collection of secondhand mirror listings where sellers tried to photograph mirrors without showing their reflection, processed through an AI image expander.
A series of interactive steps and tests you must complete before being deemed worthy enough to unlock an old casual selfie of me.
Browser-based emotion and smile tracking with Google Analytics on my personal website, inspired by corporate dystopian smile-detection cameras.
Solo performance and NYFW exhibition using marker-based augmented reality to critique fast fashion consumerism.
A diary entry that progressively obfuscates text the farther the reader is physically located from the coordinate where it was written.
Two characters in love that speak to each other across popups in the browser window.
A Node.js web app where users create customizable shrines with interactive buttons allowing others to worship them harder and harder.
A digital gift reminding a heartbroken friend that the love they put into the world remains within them.
An archive of early projects and experiments from before all this.
My day job is making demos: what can this technology do?
My art practice is making spectacles: what can this technology say?
Like a hand and a shadow puppet, they inform and motivate each other.
My medium for critique is system mimicry: a purity test that measures AI-promiscuity, an email to your ex with a 0.0001% chance of sending, a dating app that only lets you match via instagram algorithms. With each work inheriting the trust of a well-acquainted format, I onboard an unsuspicious audience to my position.
Millions of users took a quiz to locate themselves within the AI era, thousands of wistful emails have been logged into a system where almost none will ever be delivered, and a few hundred doomscrollers signed up to find their scroll-mates. Rather than offering critique at a distance, I am offering critique through participation.
And it works because the formats I mimic already have your attention. They already have your data. They already know what you do at midnight and who you looked up twice. I am just making sure you notice.
Tina Tarighian is an artist whose practice moves between demo and spectacle, repurposing the formats of contemporary technology to examine what they do and what they reveal. Working in software, installation, and internet infrastructure, she builds inside the systems she examines, producing work that functions at scale and has reached millions. She is a Senior Creative Technologist at Google Creative Lab and is based in New York.