Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I’m a Computer Science undergraduate at the National University of Singapore focusing on Artificial Intelligence and database systems. I build agentic and LLM-powered data workflows, with an emphasis on evaluation, fault localisation, and reliable execution. I enjoy turning ambiguous requirements into executable systems—whether through multi-agent text-to-SQL architectures, robust testing and CI/CD, or agentic event pipelines that automate complex engineering tasks. I’m also hands-on with front-end features, integration work, and secure, standards-driven software development.…I’m a Computer Science undergraduate at the National University of Singapore focusing on Artificial Intelligence and database systems. I build agentic and LLM-powered data workflows, with an emphasis on evaluation, fault localisation, and reliable execution. I enjoy turning ambiguous requirements into executable systems—whether through multi-agent text-to-SQL architectures, robust testing and CI/CD, or agentic event pipelines that automate complex engineering tasks. I’m also hands-on with front-end features, integration work, and secure, standards-driven software development.

Wang Xiao Qi

AI Engineer, Data Scientist, Software Engineer, +2





I’m a Computer Science undergraduate at the National University of Singapore focusing on Artificial Intelligence and database systems. I build agentic and LLM-powered data workflows, with an emphasis on evaluation, fault localisation, and reliable execution. I enjoy turning ambiguous requirements into executable systems—whether through multi-agent text-to-SQL architectures, robust testing and CI/CD, or agentic event pipelines that automate complex engineering tasks. I’m also hands-on with front-end features, integration work, and secure, standards-driven software development.…I’m a Computer Science undergraduate at the National University of Singapore focusing on Artificial Intelligence and database systems. I build agentic and LLM-powered data workflows, with an emphasis on evaluation, fault localisation, and reliable execution. I enjoy turning ambiguous requirements into executable systems—whether through multi-agent text-to-SQL architectures, robust testing and CI/CD, or agentic event pipelines that automate complex engineering tasks. I’m also hands-on with front-end features, integration work, and secure, standards-driven software development.

Available to hire

I’m a Computer Science undergraduate at the National University of Singapore focusing on Artificial Intelligence and database systems. I build agentic and LLM-powered data workflows, with an emphasis on evaluation, fault localisation, and reliable execution.

I enjoy turning ambiguous requirements into executable systems—whether through multi-agent text-to-SQL architectures, robust testing and CI/CD, or agentic event pipelines that automate complex engineering tasks. I’m also hands-on with front-end features, integration work, and secure, standards-driven software development.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Intermediate

Intermediate

Beginner

Beginner

Beginner

Beginner

Beginner

Language

English

Fluent

Work Experience

AI Research and Development at Cynapse.ai (Singapore)

July 1, 2026 - Present

Developing an agentic data system that converts natural-language requirements into executable user-defined events to automate an IOP configuration workflow that previously took months of manual engineering. Designing benchmark scenarios and evaluation pipelines for spatial, temporal, aggregate, and multi-object video-analytics conditions. Integrating the UDE generation pipeline with an existing video-intelligence platform via RabbitMQ for asynchronous inter-service communication and end-to-end event execution. Building proof-of-concept systems combining LLMs, agentic AI, and computer-vision outputs for real-world operational monitoring.

Research Project at Text-to-SQL Research Project

May 1, 2026 - Present

Designing a multi-agent text-to-SQL system integrating a model-blind planner, private semantic broker, model-specific schema agents, deterministic SQL compiler, critic, and a self-evolving meta-updater. Proposed an access-separated agent architecture leveraging opaque relation handles to isolate business-level reasoning from private schema, join metadata, data inspection, and generated SQL. Developing an interactive logical DAG to localise semantic ambiguity and execution failures to individual constraint nodes, enabling targeted user correction and selective query recompilation. Building evaluation pipelines across BIRD-Critic, Spider 2.0-Snow, BIRD-Interact, and mutation-based test suites to measure execution accuracy, fault localisation, repair success, latency, and interaction cost.

AI Intern at SINOVISION TECHNOLOGY PTE LTD (Singapore)

May 1, 2025 - August 31, 2025

Developed front-end features for Sinovision’s Ultimo IWMS platform in collaboration with UI/UX designers and backend engineers to deliver responsive, user-friendly interfaces. Built and integrated a prototype chatbot using Lang-Flow to implement simple business logic and improve user interaction workflows.

Developer (Projects) at MVC-Based Patient Management System

January 1, 2025 - December 31, 2025

Built a Java MVC-based patient management system using JUnit5, Git, and GitHub Actions for CI/CD. Implemented a high-rigour testing suite achieving 80%+ code coverage for core command logic stability and regression prevention. Enforced strict input sanitation for security and data compliance to reduce injection risks. Optimised CI/CD workflows to automate build and testing and reduce integration errors; supported team development with code reviews and technical documentation (UML, user stories).

Developer (Projects) at Dynamic Tutor Matching & Scoring Platform

January 1, 2024 - December 31, 2024

Engineered a scalable backend data model for real-time tutor-student matching, using SQL query optimisation and indexing to maintain performance under high concurrency. Integrated Stripe payments and Google Sign-In to improve security and identity linking. Collaborated on API design to translate complex location data via Google Maps into simplified user insights supporting business geographic expansion.