AI Engineer

Company Overview

Majlis CRM is a cutting-edge AI-driven real estate CRM tailored for brokerages in the UAE. The platform empowers real estate teams to efficiently manage leads, connect buyers to suitable properties, automate follow-ups, and enhance client interaction. By integrating AI directly into brokerage operations, agents can focus more on relationship-building and closing sales.

Role Summary

We seek a skilled AI Engineer to design, develop, and deploy advanced AI features within Majlis CRM. Collaborating closely with founders, product managers, and development teams, you will transform real estate workflows into AI-enhanced solutions. Projects include conversational assistants, AI-driven property searches, lead qualification systems, and call and meeting analytics. You will be responsible for taking AI features from experimentation through production, monitoring their effectiveness, and iteratively improving based on user feedback.

Key Responsibilities

• Design and implement features utilizing leading language and multimodal AI models from providers such as OpenAI, Anthropic, Google, and open-weight models.
• Create AI agents capable of retrieving data, executing functions, updating CRM records, generating tasks, and managing follow-ups with necessary permissions and human oversight.
• Build retrieval-augmented generation (RAG) pipelines involving document ingestion, parsing, chunking, embeddings, hybrid search techniques, metadata filtering, and result reranking.
• Develop natural language property search and matching systems combining semantic understanding with precise structured criteria like budget, location, bedroom count, and availability.
• Extract structured data from inquiries, conversations, and documents to aid lead classification, qualification, routing, and CRM updates.
• Implement call and meeting intelligence workflows including transcription, speaker diarization, summary generation, action item identification, and task follow-up.
• Design prompts, context handling, conversational memory, structured output formats, and tool integrations to ensure robust AI behavior.
• Establish evaluation datasets and automated testing frameworks to assess retrieval quality, factual accuracy, tool execution, and business performance.
• Maintain production AI systems to optimize response time, token consumption, operational cost, and system reliability via model selection, caching, routing, retries, and fallback strategies.
• Implement strict access controls, tenant isolation, audit logs, and protections against prompt injection and unauthorized data access.
• Assess new AI models and tools, incorporating those that deliver measurable product improvements.

Required Qualifications

• Solid foundation in software engineering with strong Python programming skills, including backend service development, REST API creation, asynchronous workflows, and database integrations.
• Proven track record of delivering and maintaining Large Language Model (LLM)-powered applications beyond basic prototypes or chatbot demos.
• Hands-on experience with major LLM APIs, such as OpenAI, Anthropic Claude, or Google Gemini, and proficiency in comparing models based on quality, latency, cost, and privacy considerations.
• Deep understanding of prompt engineering, context management, function-calling APIs, structured outputs, schema validation, and recognizing model constraints.
• Experience constructing AI agent workflows using direct model SDKs or frameworks like LangGraph, OpenAI Agents SDK, or similar.
• Expertise in retrieval-augmented generation (RAG) systems, embedding models, vector search technologies, and relational databases, using tools such as PostgreSQL with pgvector, Qdrant, or Pinecone.
• Ability to strategically select among SQL queries, traditional application logic, retrieval methods, or LLMs to solve problems robustly.
• Skilled in evaluating and troubleshooting AI systems by tracking model and tool interactions, diagnosing failures, and maintaining regression test suites.
• Familiar with development tools and processes including Git version control, Docker containers, CI/CD pipelines, automated testing, and deployment on leading cloud platforms.
• Strong analytical problem-solving capability, clear communication, and ownership of features from initial design through final production release.

Preferred Attributes

We prioritize engineers who can demonstrate operational AI products, explain their technical choices with clarity, and showcase how they assessed and enhanced feature outcomes. Familiarity with every specific tool listed is not mandatory; exceptional engineering judgment, pragmatic execution, and rapid learning aptitude are highly valued.