Archive position — measured, not model output
1 like on Devpost
506 of the 7,856 archived projects have more likes, and 1,758 share exactly 1 — so this project's #1,409 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
MaLu Estate AI Agency OS is a self-reported AI-native operating system for real-estate agencies, built by a single founder who is not a software developer. The system uses AI agents (powered by Codex models via OpenClaw) to prepare operational work such as lead follow-ups, listing copy, and internal agency briefs, while humans remain in control of trust-critical decisions. It is being developed alongside a live Croatian real-estate agency.
What changed
The project was submitted to the OpenAI 2026 hackathon. The description indicates that it is not a mockup but a working system being developed with an actual agency, and that the founder used AI agents to execute business logic without manual coding.
Single most important open question
Is there evidence of real-world traction or adoption beyond the single live agency, or any indication of how the platform will scale beyond its current scope?
What The Product Actually Is
The description states that MaLu Estate AI Agency OS is an AI Agency Command Center designed for real-estate operations. It includes three specialized agents:
- Karlo / FOLLOWUP: Reviews lead context and proposes safe next-best-action follow-ups for human review.
- Nora / COPY: Creates fact-checked listing copy from verified property data and turns unsupported claims into internal verification tasks.
- Lana / MANAGER: Produces an internal agency brief based on metrics like leads, waiting inquiries, overdue tasks, pending approvals, and listing readiness.
The system is described as a review-first workflow, where AI agents prepare drafts or escalations but do not automatically send messages, publish listings, or mutate production data. It is built using Next.js, Fastify, PostgreSQL/Supabase-style schema, TypeScript, and uses Codex models through OpenClaw for execution.
Evidence
- The description explicitly names three agents (Karlo, Nora, Lana) with defined roles.
- It describes a demo that is "production-isolated, fixture-only, and review-first."
- It mentions the use of AI agents to prepare operational work while preserving human approval points.
Inference The system appears to be an early-stage platform for managing real-estate agency workflows using AI-assisted automation, with a focus on operational efficiency and human oversight.
Positioning & Claim Evolution
The description states that MaLu Estate AI Agency OS is being built "alongside a real Croatian real-estate agency whose launch process started about 10 days before this submission." The goal was to explore what an AI-native agency OS could look like when the founder understands the market and customer workflow, but is not a software developer.
The positioning is that of a review-first AI system, where AI agents prepare work for human review rather than automating decisions. It is described as being built with "a Croatian real-estate market context in mind."
Evidence
- The project is positioned as an AI-native agency OS.
- The focus is on operational efficiency, not full automation.
- The system is localized for the Croatian market.
Inference The company positions itself as a tool that supports real-estate agents by reducing repetitive tasks while maintaining human control over trust-critical decisions. It is not a general-purpose AI platform but a domain-specific solution for real estate.
Target Customer & ICP
The description states that the system is being built "alongside a real Croatian real-estate agency." The agency is already live and operational, and the software is being developed in tandem with its needs. The system is intended to support real-estate agents in managing leads, listings, marketing, documents, tasks, approvals, and management visibility.
Evidence
- The project is built for a Croatian real-estate agency.
- It supports lead management, listing preparation, marketing, document workflows, and task coordination.
- The system is intended to grow with the agency.
Inference The ICP appears to be real-estate agents or agencies in Croatia, with potential expansion into other markets. The focus is on operational tasks rather than end-user clients.
Business Model & Pricing Evidence
Not evidenced.
Evidence
- No mention of pricing, revenue model, or monetization strategy.
- No indication of whether the platform will be sold to agencies or offered as a service.
Inference The business model is unclear. It may evolve into a SaaS or agency-based offering, but no evidence of this exists in the description.
Technical & Delivery Signals
The system is built using:
- Frontend: Next.js, React, TypeScript
- Backend: Fastify Tool API, Node.js, PostgreSQL/Supabase-style schema
- AI Execution Layer: Codex models via OpenClaw
- Infrastructure: GitHub, Supabase/PostgreSQL, Vercel, Cloudflare
The system is described as using a custom agent system and a Hermes runtime concept, with audit logs, per-agent scopes, outbox approval, and fixture-based demo data.
Evidence
- The tech stack includes Next.js, Fastify, PostgreSQL, Supabase, TypeScript.
- AI agents are used to execute business logic via Codex/OpenClaw.
- The system uses a review-first workflow with auditability and outbox approval.
Inference The platform is built on modern web and backend technologies, with an AI execution layer that allows non-developers to build software. It is designed for operational visibility and human control.
Traction & Maturity Signals
The description states that the agency and the software are both already online and live, and that the system is being developed together with the agency’s actual operational needs. The project was submitted to a hackathon, and it is described as not being a mockup but a working system.
Evidence
- The real agency is already live.
- The system is being developed alongside the agency.
- It is not a mockup or prototype.
Inference There is early traction with a live agency using the platform. However, no data on adoption, usage metrics, or customer retention is provided.
Competitive Context
Not evidenced.
Evidence
- No mention of competitors or market positioning relative to existing tools.
- No indication of how this compares to CRM systems, real-estate platforms, or AI tools in the space.
Inference The competitive context is unclear. It may compete with CRM tools or real-estate-specific platforms, but no evidence supports this.
Key Risks & Red Flags
- Single-founder model: The system is built by a single person who is not a developer.
- No revenue or customer data: No evidence of monetization, customers, or traction beyond the live agency.
- Limited scope: The demo is review-first and does not include end-to-end automation.
- Unclear scalability: No indication of how the platform will scale beyond one agency or one market (Croatia).
- Unverified claims: All evidence is self-reported, with no independent verification.
Evidence
- Only one team member listed: Nikita AI Assistant.
- No revenue, customer, or adoption data provided.
- The system is not fully autonomous and remains in a demo state.
Diligence Questions To Ask The Founders
- What are the actual operational needs of the Croatian agency that are being met by this platform?
- How does the founder validate that the AI agents are producing accurate outputs, especially for listing copy?
- Is there any plan to expand beyond the Croatian market or real-estate vertical?
- What is the long-term vision for monetization and customer acquisition?
- How is the system being tested in real-world conditions with the agency?
- Are there any plans to integrate third-party tools or APIs?
Investment/Partnership Verdict
Not evidenced.
The description does not provide sufficient information to assess whether this project is a viable investment or partnership opportunity. It is unclear if the platform has traction, scalability, or a clear path to monetization.
Evidence
- No revenue, customer, or adoption data.
- No indication of market size or competitive positioning.
- No evidence of a scalable business model.
Inference This project appears to be an early-stage prototype with real-world application but lacks the commercial signals needed for investment or partnership decisions. It may have potential if it scales beyond one agency and demonstrates measurable impact.
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.