Research & Innovation Lab

AI Research at Data Stella

We invest in applied AI research to build tomorrow's enterprise solutions today — publishing findings, open-sourcing tooling, and transferring breakthroughs directly into client engagements.

15+
Published Insights
8
Active Research Tracks
30+
AI Models in Production
3
Patent Applications

Research Areas

Six active tracks covering the full stack of modern AI — from foundation models to deployment infrastructure and responsible use.

Generative AI

Large language model fine-tuning, prompt engineering, and domain-adapted generation for enterprise use cases.

LLMsFine-tuningPrompt Engineering

Natural Language Processing

Semantic understanding, named entity recognition, intent classification, and multilingual NLP for APAC markets.

NERSentimentMultilingual

Computer Vision

Object detection, document digitization, OCR pipelines, and visual quality inspection for manufacturing and retail.

OCRObject DetectionVisual QA

Agentic AI Systems

Multi-agent orchestration, autonomous workflow execution, and tool-use frameworks for complex business processes.

AgentsOrchestrationTool Use

RAG & Knowledge Systems

Retrieval-augmented generation, vector databases, and hybrid search to ground AI on proprietary enterprise data.

RAGVector DBHybrid Search

Responsible AI

Bias detection, model explainability, data privacy frameworks, and AI governance aligned with Singapore AI guidelines.

FairnessXAIAI Governance

Our Research Methodology

Applied research built to ship — every finding is validated against real enterprise workloads and deployed into production.

01

Identify

Surface high-impact problems from client engagements and industry signals.

02

Prototype

Rapid experimentation with state-of-the-art models, techniques, and architectures.

03

Validate

Rigorous benchmarking against baselines and real-world enterprise datasets.

04

Deploy

Transfer proven techniques directly into client solutions and our product suite.

Insights & Publications

Technical deep-dives, white papers, and research briefs from our AI lab — grounded in real deployments.

White PaperJanuary 2025

Deploying RAG in Enterprise Environments: A Singapore Case Study

How we reduced hallucination rates by 94% using hybrid retrieval and re-ranking across a 2M-document knowledge base.

Research BriefNovember 2024

Fine-tuning LLMs for Domain-Specific Customer Support in APAC

Benchmark results across five vertical-specific fine-tuned models vs. general-purpose LLMs on Bahasa, Thai, and Mandarin support tickets.

Technical ReportOctober 2024

Agentic AI: Building Autonomous Business Workflows at Scale

Architecture patterns, failure modes, and production lessons from deploying multi-agent systems across logistics and finance clients.

Research Partnerships

Partner with Our AI Research Lab

We collaborate with enterprises, universities, and government agencies on applied AI research. Get early access to findings, co-develop models on your data, and accelerate your AI roadmap.