Archive position — measured, not model output
0 likes on Devpost
2,264 of the 7,856 archived projects have more likes, and 5,592 share exactly 0 — so this project's #2,646 place in the like-ranked listing is a tie-break inside that group, not a ranking.
Projects (log scale)
Likes on Devpost. ▲ marks this project's group.
Show the figures
| Likes | Projects | Share of archive |
|---|---|---|
| 0 | 5,592 | 71.2% |
| 1 | 1,758 | 22.4% |
| 2 | 285 | 3.6% |
| 3–4 | 132 | 1.7% |
| 5–9 | 75 | 1.0% |
| 10+ | 14 | 0.2% |
Executive Summary
What the company appears to be
Anastasis is a self-reported tool that claims to resurrect dead or overpriced web apps using AI. The authors describe it as a system that takes a 2-minute screen recording and an export ZIP file, then uses GPT-5.6 vision and Codex to rebuild the app with data migrated, live on a URL owned by the user.
What changed
The project is presented as a hackathon submission (Devpost entry for OpenAI 2026). No prior version or commercial product is evidenced. The authors describe it as a proof-of-concept that works end-to-end in production, including AI-driven app rebuilding, verification gates, and deployment.
Single most important open question
Is there any evidence of real-world usage, customer feedback, or traction beyond the hackathon demo? The description states no revenue, customers, or adoption data exist beyond the authors' own claims.
What The Product Actually Is
The description states that Anastasis watches a 2-minute screen recording and reads an export ZIP file to rebuild a web app using AI. It uses GPT-5.6 vision for observation, Codex for code generation, and Kubernetes-based infrastructure for deployment.
Inference This is a tool that attempts to reverse-engineer a web application from user behavior (video) and data exports, then rebuilds it with the same functionality and data, but under the user’s control.
Evidence
- “Anastasis watches a 2-min screen recording, reads your export, and Codex rebuilds the app you actually used”
- “GPT-5.6 vision watches frame batches from the recording and writes exhaustive observation notes”
- “Codex (gpt-5.6-sol) generates the entire app from scratch — schema, migration, API routes, UI”
Not evidenced
- No actual product or live URL is provided.
- No real-world use cases or customer data are described.
Positioning & Claim Evolution
The authors state that Anastasis addresses two core problems:
- Apps shutting down (e.g., due to acquisition)
- Overpriced tools where users only use 20% of features
They claim the tool flips the control dynamic — “what if your workflow could never be taken away from you again?”
Inference The positioning is that of a data-portability and user-control solution, aimed at empowering users to retain ownership over their workflows and data.
Evidence
- “Every one of us has lost an app we loved. A company gets acquired, a service shuts down, and you're left holding a ZIP file of your own data useless without the app that made it meaningful.”
- “We wanted to flip that: what if your workflow could never be taken away from you again?”
Not evidenced
- No mention of competitors or market positioning beyond self-description.
- No evidence of prior product iteration or evolution.
Target Customer & ICP
The description implies the target is users who have lost access to apps they depend on, or those who feel overcharged for software they don’t fully use.
Inference The ideal customer is someone with a valuable but dead or expensive app, who wants to regain control of their data and workflow.
Evidence
- “Every one of us has lost an app we loved.”
- “The tool that charges $30/month while you use 20% of its features.”
Not evidenced
- No stated customer segments, personas, or buyer journeys.
- No evidence of market research or early adopter feedback.
Business Model & Pricing Evidence
No pricing model or monetization strategy is described. The authors do not state whether the tool will be free, subscription-based, or offered as a service.
Inference The business model is unclear and likely under development, if at all.
Evidence
- No mention of pricing, subscriptions, or monetization.
- The tool is presented as a hackathon demo with no commercial intent stated.
Not evidenced
- No revenue streams, pricing tiers, or customer acquisition costs.
Technical & Delivery Signals
The authors describe a complex technical stack including GPT-5.6 vision, Codex, Kubernetes (k3s), Docker, Next.js, and Cloudflare. They also detail challenges such as OOM issues, sandboxing, and ephemeral pod handling.
Inference The project is technically ambitious, with a focus on AI-driven code generation and deployment automation.
Evidence
- “GPT-5.6 vision watches frame batches from the recording and writes exhaustive observation notes”
- “Codex (gpt-5.6-sol) generates the entire app from scratch”
- “Production infra: a 2-node k3s cluster across two VPSes, in-cluster registry, Kaniko image builds per resurrection”
Not evidenced
- No evidence of scalability or production stability beyond demo.
- No mention of performance benchmarks or infrastructure costs.
Traction & Maturity Signals
The project is described as a hackathon submission and is presented as a working end-to-end demo. The authors claim to have built a complete, live, verified app with 412 records migrated.
Inference This is a proof-of-concept, not a commercial product or market-ready solution.
Evidence
- “A complete, live, end-to-end resurrection in production”
- “Browser upload → AI pipeline → interactive clarification → verified build → containerized → deployed”
Not evidenced
- No customer base, usage metrics, or adoption data.
- No evidence of product iteration or feedback loops beyond the demo.
Competitive Context
No mention of existing competitors or market players is provided. The authors do not reference similar tools or platforms in the space of app resurrection or data portability.
Inference The competitive landscape is unknown and likely unexplored in this description.
Evidence
- No references to similar products, services, or platforms.
- No mention of existing solutions for data portability or app migration.
Not evidenced
- No competitive analysis or positioning against other tools.
Key Risks & Red Flags
Several technical and commercial risks are implied in the description:
- AI Reliability: The system depends heavily on AI for code generation, which may fail or misinterpret data.
- Scalability: The demo is built on a 2-node k3s cluster; no evidence of scaling beyond that.
- User Dependency: The system requires user input during ambiguous cases, which may slow adoption.
- Lack of Traction: No evidence of real-world usage or customer feedback.
Inference The tool is experimental and not yet proven in a commercial context.
Evidence
- “When the AI hits genuine ambiguity… it pauses mid-build and asks you instead of silently guessing.”
- “This project fought back, and every fix came from a real failed run.”
Not evidenced
- No evidence of user feedback or product-market fit.
- No indication of long-term viability or scalability.
Diligence Questions To Ask The Founders
- What is the actual data loss scenario you're solving for? Is this a niche use case or broader market need?
- How does the system handle ambiguous or conflicting data between the video and export?
- Have you tested with real-world apps, or is it limited to demo cases?
- Are there any legal or compliance issues around rebuilding apps that may be proprietary or copyrighted?
- What are your plans for monetization or product development beyond this hackathon project?
Investment/Partnership Verdict
Not evidenced.
The description does not provide sufficient evidence of traction, revenue, customer adoption, or a clear path to commercialization. It is a self-reported hackathon demo with no verified commercial activity.
Inference This is an early-stage idea with technical ambition but no demonstrated market readiness or business model. It may be worth exploring further if the founders can demonstrate real-world usage or traction beyond the demo.
Source
Submitted to the OpenAI 2026 hackathon on Devpost. Project home on DevPost.
The analysis above was generated by a language model from the project's own one-line description. It is not independent research and contains no verified traction, revenue or customer data.