Core Responsibilities:
- Design and build the central data infrastructure (ClickHouse, data lake, or equivalent) that aggregates data from property management, leasing, finance, and operations systems.
- Integrate frontier LLMs (Claude, GPT-4, or equivalent) via API to serve as the reasoning and analysis layer on top of internal data.
- Build and maintain retrieval pipelines that intelligently select, filter, and deliver relevant data to the LLM within context window constraints.
- Develop internal tools, dashboards, or interfaces that allow non-technical staff to interact with AI-powered analysis.
- Implement proper data security, access controls, and compliance protocols ensuring sensitive real estate and tenant data is handled appropriately.
- Architect systems with modularity in mind LLM providers, data sources, and tools should be swappable as the landscape evolves.
- Monitor AI output quality and implement validation/feedback loops to catch hallucinations and errors.
- Stay current with AI developments and proactively recommend when new capabilities become relevant to Unified's operations.
- Document all systems, architectures, and processes thoroughly. Bus-factor risk must be minimized.
Required Qualifications
- 35 years of software engineering experience, with strong proficiency in Python.
- Hands-on experience building applications that integrate with LLM APIs (Anthropic, OpenAI, or similar).
- Practical knowledge of data engineering: ETL pipelines, SQL, working with structured and semi-structured data on scale.
- Experience with at least one of: ClickHouse, PostgreSQL, BigQuery, Snowflake, or comparable analytical databases.
- Understanding of retrieval-augmented generation (RAG) patterns, vector databases, and context management strategies.
- Familiarity with orchestration frameworks (LangChain, LlamaIndex, or custom equivalents).
- Ability to deploy and maintain production services (Docker, cloud infrastructure, CI/CD basics).
- Strong communication skills you will need to explain technical constraints and trade-offs to non-technical stakeholders clearly.
Preferred Qualifications
- Experience in real estate, property management, or construction technology is a strong plus but not required.
- Exposure to agent-based AI workflows and multi-step task orchestration.
- Experience with data visualization tools or building internal dashboards (Streamlit, Retool, or custom).
- Familiarity with Arabic language NLP or bilingual data environments.
- Previous experience working in small teams or startup-like environments where you owned projects end-to-end.
What This Role Is Not
- This is not an ML research role. You will not be training or fine-tuning models. You will be implementing and integrating existing frontier models.
- This is not a pure data engineering role. You will build data infrastructure, but the goal is to power AI applications, not just maintain databases.
- This is not a management role. This is an individual contributor position with high autonomy and direct impact.
Working Arrangement
Based in Riyadh. You will participate in weekly rotating meetings with different departments (leasing, property management, finance, operations) alongside the AI Analyst to understand business needs and demonstrate progress. Expect a fast iteration cycle the expectation is working prototypes, not lengthy planning phases.
