Abdulrahman Mousa

Drive the AI feature direction and technical design for our video collaboration platform, working hands-on with engineering for private enterprise, KSA government, university, and healthcare clients across MENA. AI product strategy Defined the 12-month AI roadmap (meeting recap, smart-search, action-item extraction, agentic workflows), prioritizing against business goals and technical feasibility; AI features lifted CSAT by 18 points post-launch. Built the evaluation framework (precision, recall, groundedness, regression sets) that gates every AI release, reducing post-release defects by ~40%. Architecture Partnered with engineering on architectural direction for a microservices platform on AWS (Docker, Kubernetes), ensuring multi-tenant scalability and on-prem deployability for regulated clients. Led discovery with 25+ stakeholders across enterprise, government, university, and healthcare segments; turned findings into requirements and acceptance criteria that cut rework by ~35%. Delivery Coordinated quarterly releases across engineering, design, QA, and customer-facing teams; owned rollout and customer handoff. Tracked adoption, retention, CSAT, and feature-level metrics to re-prioritize the roadmap each quarter. Stack: Generative AI, RAG, AWS, Kubernetes, microservices, evaluation/MLOps, KPI dashboards.

Bridge role between engineering and product before transitioning into Product Management — focused on AI feature definition for regulated, on-prem customers. • Authored the BRD, acceptance criteria, and audit/logging requirements for an LLM-based smart-search module across customer support conversations — retrieval in <2 seconds while meeting on-prem and KSA data-residency requirements. • Used AI-assisted analysis (LLM summarization + topic clustering) on 300+ tickets and call transcripts to surface the top 5 user pain points; two became roadmap themes that contributed to the +18 CSAT lift. • Defined evaluation metrics (precision, recall, groundedness) and rollout gates for AI features in partnership with engineering — cut post-release defects ~40% and established the responsible-AI baseline still used across the platform.

The hands-on engineering foundation behind everything I build: shipped GenAI infrastructure, full-stack features, and CI/CD from scratch. Designed and shipped end-to-end Generative AI solutions: RAG architectures, embedding pipelines, prompt-engineering patterns, grounding/evaluation strategies, and AI agents with tool use, multi-step reasoning, and guardrails, all on a vector retrieval stack with high-availability MongoDB. Architected backend with OOP principles for extensibility; reduced response times by ~25% through profiling, refactoring, and bottleneck analysis on messaging and core communication modules. Established a complete GitLab + Docker CI/CD ecosystem from scratch (repos, branch policies, build/test/quality stages, containerized deployments), eliminating manual releases and simplifying rollbacks. Built React.js front-ends and Node.js/Express backend with WebRTC peer-to-peer video/audio, room management, and participant controls, delivering tailored, responsive client-facing apps per customer.

- Produced presales and pre-development deliverables for a founding team building an AI-driven healthcare product (solution architecture document, technical proposal, PoC plan, sizing and effort estimates, and phased execution roadmap), establishing the technical baseline. - Evaluated technology options and co-designed the core architecture, recommending a stack that balanced scalability, maintainability, and time-to-market for the MVP. - Planned development phases, rough timelines, and responsibilities, and aligned them with target segments, value propositions, and launch plans in a single execution roadmap.