# ElementX

> Production-grade enterprise AI consultancy, since 2013. Based in Auckland, New Zealand. ElementX designs, builds, and deploys bespoke enterprise AI that fits existing systems, meets governance requirements, and stays in production. Specialists in delivering for enterprise organisations in Financial Services, Insurance, Healthcare, Higher Education, Utilities, Government, Infrastructure, and Retail.

Four solution categories: Knowledge Agents, Agentic AI, Customer Assistants, and Bespoke builds. Engagements run in three phases: Strategy & Readiness, Design, Build & Deploy, and Enablement & Support. Named enterprise references include Mott MacDonald, University of Auckland, Southern Cross Health Insurance, NZ Defence Force, The Warehouse Group, and Deutsche Telekom.

## About
- [About ElementX](https://www.elementx.ai/about): Team, principles, partners, and how every build runs end to end.
- [Contact](https://www.elementx.ai/contact): Talk to us about an enterprise AI project.
- [Careers](https://www.elementx.ai/careers): Current open roles at ElementX.

## What we build
- [Knowledge Agents](https://www.elementx.ai/solutions/knowledge-agents): Enterprise AI that finds what your team needs across systems, scoped to what each user is authorised to see. Mott MacDonald's deployment serves 20,000+ staff globally, deployed in under six months.
- [Agentic AI](https://www.elementx.ai/solutions/agentic-systems): Production-grade AI that takes action across your systems, with the guardrails to do it safely. For workflows where repetitive multi-system work is consuming specialist team time.
- [Customer Assistants](https://www.elementx.ai/solutions/customer-assistants): Customer-facing AI that scales and stays reliable for years. The University of Auckland assistant has handled 70,000+ conversations in year one with 90%+ containment.
- [Bespoke builds](https://www.elementx.ai/solutions/bespoke-builds): Custom enterprise AI for problems off-the-shelf platforms can't solve: computer vision, document intelligence, forecasting, classification.

## How we work
- [Services](https://www.elementx.ai/services): The three phases of an ElementX engagement: AI Strategy & Readiness, AI Design, Build & Deploy, and AI Enablement & Support.

## Customer stories
- [All customer stories](https://www.elementx.ai/customer-stories): Production AI deployments across NZ enterprise and global engineering. Filterable by industry.
- [Mott MacDonald](https://www.elementx.ai/customer-stories/every-mott-macdonald-answer): Full deployment in six months, 3x engagement
- [The Warehouse](https://www.elementx.ai/customer-stories/gift-recommendation-engine): 28% higher average order value, 10,000+ new users
- [Southern Cross Health Insurance](https://www.elementx.ai/customer-stories/southern-cross-health-insurance): Southern Cross Health Insurance customer story.
- [Arcus Lending](https://www.elementx.ai/customer-stories/worlds-first-mortgage-lending-digital-human-assistant): Arcus Lending customer story.
- [University of Waikato](https://www.elementx.ai/customer-stories/university-of-waikato): Empowering Support Centre Staff with an Agile Virtual Assistant
- [Waipapa Taumata Rau, University of Auckland](https://www.elementx.ai/customer-stories/university-of-auckland-ai-chatbot-case-study): 90%+ coverage and containment across student queries
- [Autogrow Systems](https://www.elementx.ai/customer-stories/agtech-iot): Development partner for Autogrow, an automated growing platform using IoT, data science and AI to increase plant yield.
- [Peoply](https://www.elementx.ai/customer-stories/education-platform): Peoply customer story.
- [Cove Insurance](https://www.elementx.ai/customer-stories/cove-insurance-chatbot): Cove Insurance customer story.
- [Southern Cross Health Society](https://www.elementx.ai/customer-stories/insurance-digital-human-assistant): 7,000 conversations in three months, 95% satisfaction
- [The Warehouse](https://www.elementx.ai/customer-stories/product-labelling-automation): Automated product attribute labelling that enabled The Warehouse to launch a "shop by age" experience for toys without manual data classification.
- [vWork](https://www.elementx.ai/customer-stories/ai-document-extraction): vWork customer story.
- [iMonitor](https://www.elementx.ai/customer-stories/imonitor-helpdesk-automation): Retained around 35 customers in the first three months
- [iSell](https://www.elementx.ai/customer-stories/retail-product-categorisation-automation): 116,000+ products sorted into 1,250+ categories
- [UVLens® Sun Safety App](https://www.elementx.ai/customer-stories/uvlens-sun-safety): 500k+ users worldwide, featured by the Apple App Store

## Whitepapers
- [What is RAG](https://www.elementx.ai/whitepapers/retrieval-augmented-generation): What Retrieval Augmented Generation is, how it works, structural and security considerations, pros and cons, and use cases across industries to help you identify the best, most profitable way to use RAG in your business.
- [Tier 0 Customer Support: Best Practices in Leveraging AI for Self Service Support](https://www.elementx.ai/whitepapers/tier-0-customer-support): How AI-powered virtual assistants drive efficiency, increase customer satisfaction, and lower operational costs by helping customers help themselves with a Tier 0 support system.
- [Which Conversational AI Platform Should I Choose?](https://www.elementx.ai/whitepapers/conversational-ai-platforms): Broad review and comparison of the major conversational AI platforms (Amazon Lex, Google Dialogflow, IBM Watson Assistant, Microsoft Bot Framework), with strengths, weaknesses, pricing structures, and key considerations for each.

## Blog (recent 20 posts)
- [Notes and takeaways from the Claude Meetup Aotearoa](https://www.elementx.ai/blog/claude-meetup-aotearoa-july-2026): Notes from the Auckland Claude meetup: AI adoption at airline scale, governance as an accelerator, agents as teammates, and who the future gets built for. With commentary from the ElementX crew.
- [The Hidden Cost of Waiting on AI](https://www.elementx.ai/blog/the-hidden-cost-of-waiting-on-ai): What's the risk of doing nothing? Richard McLean explores why staying on the sidelines might be the most costly choice.
- [From Chaos to Clarity: Navigating Data Duplication, Abundant Data, and Query Optimisation](https://www.elementx.ai/blog/from-chaos-to-clarity-navigating-data-duplication-abundant-data-and-query-optimisation): A practical guide to making Retrieval-Augmented Generation models perform on real enterprise data, covering data cleaning, deduplication, and query strategies like vector search, semantic reranking, and HyDE.
- [The Importance of Explainability in AI](https://www.elementx.ai/blog/the-importance-of-explainability-in-ai): With generative AI now routine, businesses need to weigh up when a transparent 'glass box' model is the better choice over a powerful but opaque one, especially in finance, healthcare, legal, and education where decisions carry real weight.
- [Enhancing Education with RAG: How Universities Can Benefit](https://www.elementx.ai/blog/enhancing-education-with-rag): How universities can use Retrieval-Augmented Generation to keep AI responses grounded in their own course materials, improving search, personalising the learning experience, and giving students access to multimedia resources through a single interface.
- [Introducing Sophie](https://www.elementx.ai/blog/introducing-sophie): How ElementX, in partnership with UneeQ, brought a digital human named Sophie to the Aotearoa AI Summit panel as an unscripted AI participant, pulling insights from a large language model in real time.
- [Happy 10th Birthday, ElementX!](https://www.elementx.ai/blog/happy-10th-birthday-elementx): Founders Daniel Xu, Ming Cheuk, and Richard McLean reflect on a decade of building ElementX (formerly Spark 64), the AI breakthroughs they have lived through, and the lessons they pass on to organisations starting their own AI journey.
- [Assisted Search: The Natural Evolution of Search](https://www.elementx.ai/blog/assisted-search-the-natural-evolution-of-search): A walkthrough of the four levels of search (keyword, semantic, contextual, and generative) and how enterprises can move beyond simple keyword matching to give customers and staff faster access to the right answer.
- [Unleashing the Power of Search: Exploring the 4 Levels of Advanced Search for Enterprises](https://www.elementx.ai/blog/unleashing-the-power-of-search-exploring-the-4-levels-of-advanced-search-for-enterprises): Why AI-assisted search is the next step beyond traditional search engines for enterprises, with examples from healthcare and legal, plus the guardrails and disclaimers needed to manage the risk of inaccurate AI-generated answers.
- [Revolutionising Tier 0 and Tier 1 Support with AI Virtual Assistants](https://www.elementx.ai/blog/revolutionising-tier-0-and-tier-1-support-with-ai-virtual-assistants): How AI virtual assistants can power Tier 0 (self-service) and Tier 1 (basic) customer support, covering FAQs, knowledge bases, virtual assistants, and smart search.
- [Zero/Few/One-Shot Learning](https://www.elementx.ai/blog/zero-few-one-shot-learning): An introduction to zero-shot, one-shot, and few-shot learning techniques that let machine learning models work with very little training data, where each approach shines, and where it falls short of traditional deep learning.
- [Our 2023 summer internships](https://www.elementx.ai/blog/our-2023-summer-internships): A look at the projects our 2023 summer interns Fuki and Raining worked on at ElementX, from attribute extraction to generative 3D models.
- [Attribution Extraction for Retail Products](https://www.elementx.ai/blog/attribution-extraction-for-retail): How natural language processing makes it practical to pull structured product attributes (RAM, screen size, features) out of free-text supplier descriptions, helping retailers tag thousands of products and improving the on-site shopping experience.
- [The Future of Retail: How AI is Changing the Way We Shop](https://www.elementx.ai/blog/the-future-of-retail-how-ai-is-changing-the-way-we-shop): How AI is reshaping retail: personalised recommendations, conversational shopping assistants, smarter product search, and inventory management, with examples including The Warehouse Group's gift recommendation engine.
- [9 Awesome Applications of Computer Vision](https://www.elementx.ai/blog/9-awesome-applications-of-computer-vision): Nine industries where computer vision is delivering measurable ROI today: retail, manufacturing, healthcare, transportation, agriculture, banking, energy, food and beverage, and environmental monitoring.
- [Chatbot Implementation: Our CTO's Blueprint for Success](https://www.elementx.ai/blog/chatbot-implementation-our-ctos-blueprint-for-success): CTO Ming Cheuk's six-step checklist for planning a chatbot project, from defining the underlying problem and choosing a platform to training the model, monitoring metrics, and maintaining the bot once it is live.
- [AI Governance Series: To Build or to Buy?](https://www.elementx.ai/blog/ai-governance-series-to-build-or-to-buy): How boards should approach the build-or-buy question for AI, covering trade-offs around speed, control, talent, maintenance, and capex versus opex, plus the often non-binary middle ground of building on top of pre-trained models.
- [Looking Ahead to 2023: AI Developments our Founders are Watching](https://www.elementx.ai/blog/looking-ahead-to-2023): Founders Ming Cheuk, Daniel Xu, and Richard McLean on the AI developments they were watching for the year ahead: foundation models, generative AI in production, AI in defence, and the democratisation of access for smaller organisations.
- [AI Governance Series: Keeping regulation and compliance concerns front of mind during AI project planning](https://www.elementx.ai/blog/ai-governance-series-keeping-regulation-and-compliance-concerns-front-of-mind-during-ai-project-planning): As AI continues to become more widespread in our daily lives, it is important to consider the regulatory and compliance requirements associated with such projects.
- [AI Governance Series: Preparing for and Avoiding Roadblocks to Adoption](https://www.elementx.ai/blog/avoiding-roadblocks-to-adoption): Part three of the AI Governance series. Three roadblocks that quietly kill AI projects: data preparedness, organisational culture, and change management, plus the questions a board should ask before greenlighting a project.
- [All posts and archive](https://www.elementx.ai/blog): Full blog index with categories and search.

## Team
- [Carol Cheng](https://www.elementx.ai/team/carol-cheng): CFO
- [Carter Apas](https://www.elementx.ai/team/carter-apas): Senior Full Stack Engineer
- [Daniel Jimenez](https://www.elementx.ai/team/daniel-jimenez): Engineering Operations Lead
- [Daniel Xu](https://www.elementx.ai/team/daniel-xu): CEO & Co-founder
- [David Kelly](https://www.elementx.ai/team/david-kelly): Independent Director
- [Emma Roberts](https://www.elementx.ai/team/emma-roberts): Operations Manager
- [Garab Dorji](https://www.elementx.ai/team/garab-dorji): Full Stack Engineer
- [George Qiao](https://www.elementx.ai/team/george-qiao): Senior Full Stack Engineer
- [Jay Oh](https://www.elementx.ai/team/jay-oh): Junior Full Stack Engineer
- [Kat Klebenow](https://www.elementx.ai/team/kat-klebenow): Head of Marketing
- [Kevin Adams](https://www.elementx.ai/team/kevin-adams): Conversational AI Designer
- [Ming Cheuk](https://www.elementx.ai/team/ming-cheuk): CTO & Co-founder
- [Niclas von Ahsen](https://www.elementx.ai/team/niclas-von-ahsen): Professional Services Lead
- [Richard McLean](https://www.elementx.ai/team/richard-mclean): Exec Chair & Co-founder
- [Simona Turin](https://www.elementx.ai/team/simona-turin): Future Director
- [Yunu Choi](https://www.elementx.ai/team/yunu-choi): Full Stack Engineer

## Optional
- [Privacy policy](https://www.elementx.ai/privacy-policy)
- [Terms and conditions](https://www.elementx.ai/terms-and-conditions)
- [Full content dump](https://www.elementx.ai/llms-full.txt): All blog posts, customer stories, services, and whitepapers concatenated as markdown for efficient ingestion.