Insights
Practical perspectives on GenAI, automation, and hiring — from the Oxydata team.
27 articles
A head-to-head look at SAP SuccessFactors and OPAL — what each does best, where each falls short, and how cost and CV-screening speed compare for enterprise hiring teams.
Recent articles
Fahmi Fadzil says AI could create 300,000–500,000 Malaysian jobs. Talent Corp warns nearly 700,000 roles need reskilling. A closer look at the numbers, data-centre boom, and what “economic value” actually means.
NVIDIA’s Hugging Face Acquisition: What It Means for Enterprise AINVIDIA’s US$12.9 billion Hugging Face acquisition could reshape open-source AI. Here’s what enterprises should know about neutrality and vendor lock-in.
How we built Agent Smith, OPAL’s guided agentic JD reviewer — and why analyse-broadly, remediate-narrowly beats one-shot rewrites and full autonomy for hire-critical content.
When should a product use guided screens, and when does an AI agent add real value? Lessons from building OPAL across recruitment workflows, CV scoring and mobile access.
A practical discovery-to-pilot loop for Malaysian enterprises and SMEs — one use case at a time, with human-in-the-loop decisions. Aligned with CRISP-DM, NIST AI RMF, and modern MLOps practice.
A practical checklist for Malaysian enterprises and SMEs choosing an AI consulting partner — strategy vs build, data readiness, PDPA, human-in-the-loop, red flags, and ten questions for the first meeting.
Digital Minister Gobind Singh Deo says MyGOV is being enhanced with agentic AI to simplify government services. A short take on what that means — and why “agentic” should mean more than a chatbot.
A summary of the Ecosystm / Red Hat / NAIO whitepaper on SME AI adoption in Malaysia — key stats on pilots vs scale, talent and cost barriers, open source interest — plus Oxydata’s take on what to do next.
Vibe coding ships prototypes in minutes — but without an 80% right database schema, AI-assisted apps hit an invisible cliff of doom loops, context rot, and technical debt.
GenAI demos look fine until they meet missing fields, conflicting definitions, and undocumented feeds. A practical data quality checklist for Malaysian enterprises — and how StringRay’s six audits de-risk AI kickoff.
A practical guide to AI consulting firms in Malaysia in 2026 — who helps with strategy, GenAI implementation, data readiness, and enterprise rollout. Where Oxydata Software fits.
Malaysian enterprises still stall on 12–18 month EDW programmes. The 2026 playbook is different: domain-first SQL Server datamarts, AI-assisted delivery, and a quality gate before you build.
A practical guide to building agentic AI for Malaysian enterprises — vectorized knowledge with pgvector, LangGraph orchestration, composable tools, token cost management, and observability from IRIS and Oxa in production.
PwC's 2026 AI Jobs Barometer finds roles that amplify human judgment — including recruiters — growing far faster than jobs where AI mainly automates routine tasks. A useful signal for Malaysian HR and hiring teams.
Every vendor claims agentic AI, but most products are workflows with an LLM inside. Learn the real difference, five Malaysian business examples, and the one litmus test to ask your vendor.
Every CV uploaded to US-hosted recruitment SaaS may breach PDPA Section 129. Learn what PII is at risk, where popular platforms store data, your compliance options, and 7 questions to ask your vendor today.
TechTalent, a licensed recruitment agency in Malaysia, used OPAL to stop volume-led shortlists — saving 70% of screening time, improving CV quality by 85%, and rebuilding hiring manager trust with structured candidate summaries.
An honest breakdown of 12 AI recruitment platforms across four categories — enterprise HCM, standalone ATS, AI CV scoring, and AI interview tools. Strengths, weaknesses, and who each is really built for. Written for enterprise HR teams in Malaysia.
When vendor volume lands on a technical lead's desk every week, good candidates get missed and hiring slows down. How BMW Credit Malaysia used OPAL to enforce quality before human review.
Traditional AI (classical ML / statistical learning) predicts from tables. Generative AI (LLMs) writes and converses. Here is how they differ — plus real Malaysian use cases for banks, retail, WhatsApp, hiring, and manufacturing.
The 10 most active AI companies in Malaysia in 2026 — recruitment, education AI, computer vision, drones, and enterprise automation. Who's building real solutions.
Digital Minister Gobind Singh Deo says the National AI Action Plan 2026–2030 — targeting 30,000 AI professionals by 2030 — is expected to launch in the third quarter after Cabinet approval.
If you work in HR or recruitment in Malaysia, someone has already asked whether AI will take your job. The honest answer is not a clean yes or no — it depends on what your team actually does and how leadership treats AI.
Most Malaysian college websites fail prospective students at the first enquiry. RAG-powered AI admissions agents answer qualification questions, fees, and booking requests instantly — 24/7 in Bahasa Malaysia or English.
For many Malaysian SMEs, WhatsApp is where business happens — but manual replies create bottlenecks as volume grows. A well-designed WhatsApp AI agent can answer common questions instantly, qualify leads, and free your team for work that needs a human.
Most organisations and recruitment agencies are still running a hiring process built for 2005. In 2026, that's not just inefficient — it's costing you money, time, and your best candidates.