O

OpenArt AI

AI art platform using Stable Diffusion

Product Data Scientist - Growth

Full-TimeUpdated on 10/20/2026
No salary listed
Mid, Senior
San Carlos, CA, USA
HybridHybrid work is allowed; remote may be considered for exceptional candidates.
Data & AnalyticsData Science
H1B Sponsorship Available

About the job

Requirements
  • Three to six years of experience in data science or analytics supporting product, growth, or lifecycle teams.
  • Experience designing, running, and analyzing many A/B tests.
  • Strong SQL and Python skills, with the ability to work with data end-to-end independently.
  • A solid statistical foundation.
  • Product intuition and insight-generation ability.
  • Ability to communicate data insights clearly to product managers, engineers, and executives.
  • High ownership and comfort working in ambiguity.
Responsibilities
  • Design, run, and analyze A/B tests and other causal experiments across signup, onboarding, activation, monetization, and retention surfaces.
  • Set experimental standards, including test sizing, minimum detectable effect, run duration, primary and secondary metrics, and readout standards.
  • Debug and resolve test issues such as sample ratio mismatch, novelty effects, interaction effects, multiple comparisons, and segment heterogeneity.
  • Build and evolve the experimentation practice through playbooks, guardrails, and post-launch validation.
  • Analyze the growth funnel, including signup, activation, paid conversion, retention, and resurrection.
  • Identify drivers of long-term retention and churn, and translate findings into product recommendations.
  • Conduct lifetime value and cohort analyses to inform pricing, packaging, and monetization decisions.
  • Perform behavioral segmentation to identify important user segments and drivers of their behavior.
  • Quantify the potential impact of proposed product changes before engineering builds them.
  • Define and own key product metrics, including activation, engagement, retention, and monetization, alongside product managers.
  • Build and validate dashboards for leaders.
  • Keep metric definitions consistent across product, marketing, and finance.
  • Translate analyses into recommendations that product managers, engineers, and executives can act on.
  • Challenge experiments that are underpowered, mis-scoped, or measuring the wrong outcome using data.
Desired Qualifications
  • Experience as an early hire at a consumer startup.
  • Experience with consumer products, marketplaces, or subscription businesses.
  • Experience with experimentation platforms such as Statsig or Optimizely.
  • Familiarity with causal inference methods beyond A/B tests, including difference-in-differences, synthetic control, propensity matching, and CUPED.
  • Experience with BigQuery, Amplitude, Metabase, Stripe, or Google Cloud Platform.
  • Exposure to artificial intelligence, machine learning, or generative AI products.
  • Comfort working across time zones with a globally distributed team.

About the company

O

OpenArt AI

View
Website

OpenArt AI offers a web platform that generates AI art using Stable Diffusion, helping users create personalized digital artwork without technical skills. It provides a user-friendly interface to generate art directly or explore a library of up to 10 million prompts, with presets to refine results, all on a freemium plan. The service differentiates itself with a large, searchable prompts database and an emphasis on ease of use, plus a community on Discord where users share work and presets. Its goal is to make AI-generated art accessible and easy for people of all skill levels to turn ideas into visual content quickly.

Company Size

201-500

Company Stage

Series A

Total Funding

$35M

Headquarters

San Francisco, California

Founded

2022

Get referred to OpenArt AI

See people who can refer or advise you

Simplify Jobs

Simplify's Take

What believers are saying

  • January 2026 Series A raised $30 million from Canaan, Basis Set, DCM, Felix.
  • June 2026 Director and September 2026 Arena broaden monetization beyond image generation.
  • NovaLand's July 2026 partnership validates OpenArt for commercial children's animation.

What critics are saying

  • ByteDance controls Seedance 2.5, squeezing OpenArt margins on video generation.
  • Hollywood and advertiser backlash escalates after AI Personality of the Year awards, 2026.
  • A platform lockout from Stable Diffusion, ByteDance, or licensors would crush OpenArt.

What makes OpenArt AI unique

  • OpenArt Arena, launched September 15, 2026, ranks models by creative task, not generic scores.
  • Director, launched June 26, 2026, generates five-minute videos through conversation alone.
  • Eight million monthly active users give OpenArt training data and distribution.

Help us improve and share your feedback! Did you find this helpful?

Benefits

  • Hybrid Work Options
  • Remote Work Options
  • Company Equity

Growth & Insights and Company News

Headcount

6 month growth

11%

1 year growth

15%

2 year growth

-12%
SmarToolbox
Sep 16th, 2026
The best AI model is the one that fits your workflow.

The best AI model is the one that fits your workflow. Generic leaderboards are useful for orientation. They are not a substitute for testing the work you need to finish. OpenArt just launched a leaderboard for creative AI models that asks a much better question than "Which model is best?" Its Arena compares image and video outputs through blind evaluations organized around jobs such as advertising, film, animation, product imagery, editing, and lip sync. OpenArt says the launch includes 28 expert judges and 1,003 tastemakers. The point is not to produce one grand winner. It is to help a creative person find the model that performs for a particular brief. That sounds like a small change in presentation. It is not. The AI tools market has spent years teaching SmarToolbox to compare models as if they were smartphones: one ranking, one score, one winner. But most people do not need "the best model." They need a model that can follow a character across shots, preserve product details, return a usable research brief, or turn a rough idea into something they can actually publish. The useful unit is moving from the model to the job. A leaderboard becomes useful when it resembles your Tuesday. OpenArt's own explanation gives a good example. A model can make a beautiful image and still get the product wrong. A video can look cinematic while ignoring the requested camera direction. A talking character can lose its identity between shots. Those are not benchmark footnotes. They are failed deliverables. If you are making an ad, "looks good" is not enough. The output has to respect a product brief, keep the logo or packaging stable, and survive the next edit. If you are building a YouTube channel, the best model may be the one that gives you consistent characters and workable motion, not the one with the prettiest single frame. That is why task-specific evaluation is more useful than a universal score. It forces the test to include the annoying details that make work real. The same idea is showing up outside creative tools. The APEX-Agents research benchmark was built from work by investment banking analysts, management consultants, and corporate lawyers. Its 480 tasks are spread across 33 simulated worlds, and professionals estimated that the tasks would take 11 to 22 hours to complete. Agents have to work across files and applications instead of answering a tidy question in a blank chat window. The headline result is almost humbling: the leading system scored 24% on Pass@1. That is not a sign that the benchmark failed. It is a sign that realistic work is harder than a clean prompt suggests. For a solo creator, the lesson is simple: test the thing you actually do. Give an AI tool your real brief, your real source material, your real formatting constraints, and your real definition of "done." A model that wins a public leaderboard but creates cleanup work may be the wrong model for you. The workflow around the model decides more than the model name. The Pragmatic Engineer's reporting on Codex describes OpenAI using the coding agent beyond ordinary code completion, with automated feedback loops and a growing need to rethink review. OpenAI's own cookbook shows the pattern in a smaller, clearer form: review an artifact, repair it, validate it, then use the result as the input to the next pass. That loop matters because AI output is cheap. Reliable output is not. A model can draft an article, generate a script, or write a pull request in seconds. The expensive part is checking whether the result obeyed the brief. Did it preserve the important facts? Did it break the build? Did it invent a source? Did it quietly change the tone? Did it create a file that looks finished but cannot be used? A good workflow makes those questions executable. It adds a checklist, a test, a linter, a blind comparison, or a human review step. The model is still important, but it is no longer the whole product. This is also why agent demos can be misleading. The impressive part is often the visible action: the agent opens a browser, edits a document, or calls a tool. The durable advantage comes from the less glamorous layer around it: permissions, state, feedback, and a way to detect failure. For my own projects, I would rather have a slightly less capable model inside a workflow that catches mistakes than a stronger model operating with no receipt for what it changed. Memory is part of the test, not a bonus feature. Long-running workflows expose another weakness in model comparisons: the model may perform well on a fresh prompt but lose the plot after a week of work. The answer is not always a larger context window. A March 2026 paper called Memori tested a persistent memory layer that converts conversations into structured facts and summaries. On the LoCoMo benchmark, it reported 81.95% accuracy while using an average of 1,294 tokens per query, about 5% of the full context in its comparison. The paper also reports more than 20 times lower token use than passing the entire conversation history. Those are results from one research system, not a universal promise. But the architectural point is easy to understand: memory is a data problem as much as a model problem. If an AI assistant helps you manage content, it needs to remember the difference between a source you verified and a claim you rejected. It needs to know which tools you already tested, which tone you prefer, and which tasks still require your approval. Dumping every previous conversation into the next prompt is a poor substitute for keeping that information in a structured, inspectable form. So add memory to your evaluation. Ask the tool to continue a project after a few days. Change one requirement. Introduce an old draft and a newer correction. See whether it retrieves the right context or confidently resurrects something you decided not to use. Build a tiny benchmark before you switch models. You do not need a research lab to compare AI tools properly. You need a small set of tasks you repeat often. For a creator, that might be turning a source pack into a clean article outline, rewriting a paragraph without changing its claims, producing five usable hooks for one audience, generating a thumbnail concept that keeps the same subject across variations, or updating a published draft while preserving links and formatting. Run each task through two or three tools. Blind the outputs if you can. Score them on the things that create downstream work: factual accuracy, instruction-following, consistency, edit time, cost, and whether you would publish or ship the result. Do not give every category equal weight. If factual mistakes are expensive, accuracy should dominate. If you are making short videos, character consistency may matter more than a model's general image score. If you are routing an agent through paid APIs, correction burden belongs in the cost calculation. The score is not meant to be scientific. It is meant to stop you from choosing based on launch-day excitement. The best AI model is not the one with the highest general score. It is the one that produces the most useful result inside your workflow, with the least correction and the clearest path to checking its work. My take: stop asking which model wins. Model rankings still have a place. They are a good way to build a shortlist, discover a new tool, or notice that a category is moving quickly. OpenArt Arena is useful precisely because it narrows the question from "best creative model" to "best fit for this kind of creative work." But a leaderboard should be the beginning of your test, not the conclusion. AI products are competing on workflow management now: background execution, memory, tool use, evaluation, and safe handoff. That creates an opening for individuals and small teams. You can build a better system without training a better model, simply by defining the task clearly and measuring the result honestly. This week, take one recurring task and make a five-case benchmark for it. Keep the prompts, inputs, outputs, and scores. Run it again when a new model arrives. That small habit will tell you more about an AI tool's value than another hour spent scrolling through a leaderboard.

Yahoo Finance
Sep 15th, 2026
OpenArt launches Arena, AI model benchmark built by creative experts for filmmakers and designers

OpenArt AI has launched OpenArt Arena, a new benchmark for AI image and video models designed specifically for creative professionals. Unlike traditional technical benchmarks, the leaderboard will be based on expert evaluations from a "Creative Expert Council" comprising industry leaders, practitioners, and top platform users, including Emmy-winning animation director William Lau. The benchmark features category-specific rankings for industries such as advertising, filmmaking, graphic design, and animation. Experts will conduct blind, side-by-side evaluations of model outputs based on criteria including aesthetics, prompt adherence, realism, and motion quality. OpenArt AI serves over 8 million monthly active users and positions Arena as a resource for creators seeking guidance on selecting appropriate AI models for specific creative tasks. The leaderboard is available at openart.ai/arena.

Yahoo Finance
Sep 12th, 2026
OpenArt Director ranks No. 1 in anime on Physion Labs' ARC 1.0 benchmark at 58% lower cost

OpenArt Director has been ranked number one in anime on Physion Labs' ARC 1.0 AI Video Agent Benchmark. The AI video generator from OpenArt AI, which serves over 8 million monthly active users, topped all eight human-preference metrics in the evaluation. The benchmark tested AI video agents on 100 prompts, generating 800 videos assessed across 19 metrics. Director led in 11 of those metrics, including emotional intent, composition, pacing, and motion quality. Director also demonstrated cost efficiency, operating at approximately 15% lower cost than the second-ranked Anime Agent 1.0 and 58% lower cost than third-placed InVideo Agent One. The evaluation examined how well agents could transform complete screenplays into coherent video stories through end-to-end production.

DreamClip
Aug 9th, 2026
Best place to access Seedance 2.5: Dreamclip vs. Competitors.

Best place to access Seedance 2.5: Dreamclip vs. Competitors. ByteDance's flagship **Seedance 2.5** has officially set the standard for high-end AI video creation, offering native 30-second generations, extreme motion consistency, and advanced multi-reference control options. But with so many platforms hosting AI models, where should you go to use it? If you are looking for the **best place to access Seedance 2.5**, this article compares **Dreamclip** directly against direct competitors (like Higgsfield, OpenArt, and InVideo) to show you why creators are choosing its pay-as-you-go workflow. Direct comparison: Dreamclip vs. Competitors. Here is a side-by-side breakdown of the user experience, pricing structure, and access capabilities of Dreamclip compared to typical subscription-based platforms: Core advantages of accessing Seedance 2.5 on Dreamclip. 1. Simple pay-as-you-go coin model (starting at ₹100). Other platforms require a high upfront commitment, asking for $30 to $100 per month just to access their video generators. On Dreamclip, you can get started for just ₹100. Its coin-based deduction is simple: a ₹100 top-up adds 100 coins to your wallet. You only spend what you use, when you need it. 2. Zero credit expiration. On subscription platforms, your monthly credits reset. If you don't use them, you lose them. On Dreamclip, the coins you buy stay in your account forever. If you top up today and generate a video six months later, your coins will be waiting for you. 3. Local INR pricing & No conversion fees. If you buy from foreign SaaS platforms, you are billed in USD, which means paying expensive bank conversion markup and international card fees. Dreamclip supports direct INR (₹) transactions, meaning you pay exactly what you see with no hidden conversion costs. 4. No tier locks on advanced features. Competitors frequently lock premium options (like 30-second outputs, 720p resolution, or multi-reference video editing) behind their highest-tier subscription plans. On Dreamclip, there are no tier locks. Every premium model - including Seedance 2.5 - is fully open to everyone from day one. Why Dreamclip? (The ultimate AI video workspace). Dreamclip isn't just an interface; it's a unified platform built to solve all your creative bottlenecks: * **No Model Restrictions (Every premium model, open to everyone):** While other platforms lock their newest and most advanced models behind expensive high-tier subscription plans, Dreamclip gives you complete access to every generator from day one. No gatekeeping, no premium tier restrictions. * **One Playground (Access the latest video and image models in one place):** Generate high-fidelity videos and photorealistic images from a single, user-friendly workspace. Switch between models like Seedance 2.5, Kling 3.0, and Grok instantly, track all creations in a 7-day history panel, and get automatic coin refunds within minutes if any generation fails. * **Pay As You Go (Start generating with just ₹100):** Access the market's most competitive rates across leading models with a starting balance of just ₹100. Its coin-based deduction is simple: a ₹100 top-up adds 100 coins to your wallet. If a model like Veo 3.1 Lite costs 20 coins, only 20 is deducted - giving you up to 5 generations. Pay only for what you use, when you need it. * **Dedicated Support (Real-time help for your creative workflow):** Get instant assistance directly through the in-playground chat bubble, active from 8:00 AM to 7:00 PM (IST) daily. For off-hours queries, reach DreamClip at [email protected] and its team will get you back on track as quickly as possible. Start generating on Dreamclip today. Accessing Seedance 2.5 doesn't need to be expensive or complicated. 1. Go to [dreamclip.in](https://dreamclip.in). 2. Top up your balance with a simple ₹100 coin purchase. 3. Open the Seedance 2.5 playground and enter your prompt. 4. Hit **Generate** and bring your ideas to life! Ready to try AI video generation? Start creating stunning videos with DreamClip today.

AI Blog First
Aug 7th, 2026
OpenArt Seedance 2.5 (2026): features, pricing & coupon code.

OpenArt Seedance 2.5 (2026): features, pricing & coupon code. By Nisha / August 7, 2026 Seedance 2.5 is ByteDance's latest AI video model, and OpenArt is one of the platforms where you can actually use it. If you've been trying to figure out what sets it apart from Seedance 2.0, or whether it's worth generating your next project with, you're likely looking for specifics rather than marketing language. This article covers what OpenArt Seedance 2.5 actually does: how the 30-second native video generation works, how the 50-reference system keeps characters consistent across a scene, and how native audio syncing changes the output compared to earlier models. If you're testing AI video tools for ads, short films, or social content, understanding these features first will help you decide if Seedance 2.5 fits your workflow before you spend any credits. What is OpenArt Seedance 2.5? OpenArt Seedance 2.5 is ByteDance's next-generation AI video model, integrated into OpenArt's video generation suite. As of this writing, it is listed as "coming soon" on OpenArt, with early access available through a waitlist. The model's defining feature is that it generates a complete 30-second video clip in a single pass, rather than stitching together several shorter clips. It supports both text-to-video and image-to-video generation, and it's built to handle full scenes - an opening shot, a middle action, and a closing frame - as one continuous output instead of separate generations pieced together afterward. Save 15% on OpenArt with Code NISHA15. Use coupon code NISHA15 at checkout to get 15% off your OpenArt subscription and start creating with Seedance 2.5 today. OpenArt Seedance 2.5 coupon code & current discount offers. Use coupon code NISHA15 at checkout for 15% off. This is in addition to two other overlapping offers currently running: * A limited-time launch promotion: new users who join the Seedance 2.5 waitlist and upgrade to a Wonder Monthly or Annual plan by August 7, 8:00 AM PT receive 10 days of unlimited Seedance 2.5 generations once the model goes live. * A standing plan discount: annual billing saves up to 27% versus paying monthly, and OpenArt is offering up to 40% off top models (including Seedance 2.5) for upgrades made by the same August 7 deadline. Check at checkout whether the coupon code stacks with the plan-level discounts, since combined savings can vary by plan and billing cycle. Key features of OpenArt Seedance 2.5. * 30-second one-shot generation. A full scene renders in a single native pass, with characters, lighting, and motion staying consistent from the first frame to the last. * Up to 50 multi-modal references. You can upload a mix of up to 50 images, video clips, or audio files to anchor characters, styles, or products across the generation. * Region-level editing. You can re-render a specific part of a clip - a face, a product label, a background - without regenerating the whole scene. * Improved prompt accuracy. OpenArt reports a 20% improvement in prompt adherence over Seedance 2.0, meaning complex briefs with specific camera moves and character actions come out closer to what you described. * Near real-time generation speed, a notable jump from the multi-minute wait times of earlier models. * Native audio sync. Sound is generated in the same pass as the video, so footsteps, ambient noise, and material sounds are timed to on-screen action rather than added afterward. OpenArt Seedance 2.5 pricing plans explained. Seedance 2.5 access runs through OpenArt's standard subscription tiers, which use a shared credit system across all of OpenArt's models. There's also a Seedance 2.5 10-Day Unlimited add-on, available to Plus subscribers for $15/month, which includes 5,000 extra credits. on whether Seedance 2.5 access is limited to specific tiers once it fully launches, as this can change post-release. How to redeem the OpenArt Seedance 2.5 promo offer. * Visit the Seedance 2.5 page on OpenArt and join the waitlist. * Upgrade to a Wonder Monthly or Annual plan and Use Coupon code NISHA15 and get 15% off. * Once Seedance 2.5 launches, your account is automatically credited with 10 days of unlimited generations. The offer is tied specifically to the Wonder tier - upgrading to Starter, Plus, or Pro won't trigger it. OpenArt Seedance 2.5 vs Seedance 2.0: what's new. The clearest upgrades from Seedance 2.0 to 2.5 are length, reference capacity, and editing control. Seedance 2.0 generated shorter clips (roughly 5-10 seconds) that often needed to be stitched together for longer content. Seedance 2.5 handles a full 30-second scene natively. Reference inputs also jumped significantly - Seedance 2.0 supported a handful of references, while Seedance 2.5 accepts up to 50, useful for scenes with multiple characters or products. Region-level editing is new to 2.5 as well, letting you fix one element without re-rendering the entire clip. How to use OpenArt Seedance 2.5 (step-by-step). * Step 1: Choose your mode. Open OpenArt's video tools, select Seedance 2.5 from the model menu, and choose Text to Video or Image to Video depending on your starting point. * Step 2: Write your prompt and load references. Describe the scene clearly, including subject, camera movement, and pacing. Upload reference images, style clips, or audio files as needed. * Step 3: Generate, edit, and export. Generate the clip, use region editing to fix specific elements if needed, then export as an MP4 with native audio included. Best use cases for OpenArt Seedance 2.5. * 30-second ads in one pass. The single-generation format keeps product appearance and lighting consistent across the opening, product moment, and closing frame. * Multi-character scene production. With 50 reference slots, you can assign a reference image to each character and maintain consistent faces across the full clip. * Product and spatial visualization. Seedance 2.5 can follow a 3D blockout mesh to pre-stage camera position and object proportions, useful for product previews and architectural walkthroughs. * Iterative campaign production. Region editing reduces wasted generations, since you can fix a specific detail instead of re-rolling the whole clip - useful for teams producing many variants. OpenArt Seedance 2.5 faqs. What is Seedance 2.5? It's ByteDance's AI video model, available through OpenArt, built around 30-second native generation, up to 50 references, and native audio sync. What generation modes does Seedance 2.5 support? Text-to-video and image-to-video. How is Seedance 2.5 different from Seedance 2.0? Longer native generation length, more reference inputs, region-level editing, and better prompt adherence. How does Seedance 2.5 handle audio? Audio is generated in the same pass as the video, so sound is synced to the visuals without a separate step. Does Seedance 2.5 support multiple input modes? Bottom line. OpenArt Seedance 2.5 is built for longer, more consistent AI video generation than its predecessor, with meaningful upgrades to reference handling and editing control. There's no dedicated coupon code right now, but the current Wonder-tier launch offer and standing plan discounts are the ways to get it at a reduced cost before August 7. If you're already considering a Wonder subscription, the timing lines up; if you're on a lower tier and mainly experimenting, it may be worth waiting to see how access unfolds after the full launch.