Tamr’s Anthony Deighton: Integrating AI into Enterprise Data

We spoke with Anthony Deighton, General Manager of Data Products at Tamr, about the company’s AI-first approach
AI Magazine speaks with Anthony Deighton, CEO at Tamr, about the power of AI and how it can be harnessed to transform data

During a time where AI continues to make its mark on the global business landscape, the importance of clear data has never been more essential. 

The power of AI is only continuing to disrupt multiple key industries. From the perspective of business MDM (master data management), AI can help to improve data accuracies, unify data from diverse sources and make accessing and analysing data simpler for all stakeholders across an organisation.

We spoke with Tamr CEO, Anthony Deighton, about the company’s AI-first approach and how data teams can harness the power of AI to use MDM within their businesses.

Talk to us about Tamr and why the company holds an AI-first approach.

Tamr’s data-centric AI specialises in wrangling messy data from various sources and transforming it into clean, consistent, and analysis-ready datasets. This is particularly crucial for the creation of golden records, which is a single, unified view of an entity (like a customer, product, or supplier) across an organisation.

Key facts: What sets Tamr apart
  • AI at the Core: Tamr's approach is fundamentally different. For over a decade, we’ve been developing and refining AI, not just basic data manipulation techniques. This translates to proven effectiveness – we hold over 16 patents on our innovative technology.
  • Speed and Accuracy: Leveraging AI allows Tamr to deliver results faster and with higher accuracy. Our customers see quicker project completion times, lower costs, and most importantly, reliable data to drive real business value.

We’ve spent the last 11 years focused on using AI and machine learning (AI/ML) to tackle the hard problem of performing accurate, enterprise data at scale. With 18 AI patents, our technology has been proven in the open market over scores of customer engagements with some of the most recognisable brands in the world. 

What are some of the ways Tamr implements AI?

Data Testing: Tamr's data quality capabilities leverage a powerful combination of reference data, pre-configured business logic, and large language models (LLMs) to handle diverse source data. This means you can standardise raw inputs with attribute-specific testing, eliminating the need for repetitive custom work for each new data source.

Record matching: We combine powerful ML techniques like random forests and vector embeddings with reference data and human-configurable rules. This ensures accurate linking of records, even with diverse profiles.

Analytical enrichment: Our analytical enrichment provides a "best view" of every entity across all sources. Integrated enrichment sources and AI-based classifiers fill in missing values, resulting in complete data sets for confident decision-making.

Data classification: Tamr's data classification functionality allows you to establish a clear and consistent structure. This facilitates efficient data organisation and categorisation, ensuring data integrity, security, and compliance.

ID persistence: ID persistence empowers you to understand your data's lineage and document its provenance. Customers gain valuable insights into their data origin and authenticity. Additionally, versioning capabilities enable you to track and manage changes over time. Access previous iterations and compare differences with ease.

Youtube Placeholder

What are the benefits for organisations that switch from traditional MDM to AI-powered data mastering?

Traditional MDM solutions don’t scale, require manual human effort to configure, curate, rely on centralised control for governance and management, and are also built for static data.

An AI-powered approach to data mastering speeds the discovery, enrichment, and maintenance of trustworthy golden records that organisations need to accelerate growth. AI delivers all the value and benefits that rules-based MDM simply cannot achieve.

By combining embedded similarity with human feedback, it achieves best-in-class match rates with external data, ensuring data accuracy and reliability.

Likewise, AI is tailored to the consumer. It integrates every identifier in the system of record with human validation, creating a personalised, single view of each customer. This approach ensures that customer interactions are informed and relevant.

Finally, the effectiveness of AI-enhanced solutions increases with use. They continuously learn and improve from machine-generated feedback, making data management more efficient and adaptive over time. This learning capability ensures that the system evolves to meet changing business needs and data landscapes.

During a time of mass AI adoption, how important is human collaboration for data teams to integrate AI-powered solutions effectively?

While AI can handle a lot of the heavy lifting, it's not infallible. Algorithms might struggle with data that is exceptionally noisy, ambiguous, or complex. That’s why human refinement is critical. Humans apply judgement and domain expertise to review and refine the AI. This could involve correcting errors, making judgement calls on ambiguous cases, or providing additional context that the AI might not have considered.

This human involvement from data teams is crucial for ensuring the highest level of accuracy and reliability in golden records. It combines the best of both worlds: AI's efficiency and scalability with human intuition and expertise.

The refinement process requires businesses to leverage both AI capabilities and human expertise, underscoring the need for solutions that are sophisticated yet user-friendly to manage and utilise this AI-enhanced data effectively.

Organisations clinging to traditional MDM risk falling behind. These solutions are slow, manual, and struggle to handle the ever-increasing volume and variety of data. This translates to inaccurate data, limited scalability, and an inability to adapt to changing data landscapes. 

AI-powered solutions flip the script. By automating tasks, leveraging ML for accuracy, and building a scalable architecture, our AI delivers faster time-to-value, superior data quality, and the flexibility to handle dynamic data. 

This empowers organisations to unlock the full potential of their data, gain a competitive edge, and achieve superior business outcomes.

Moving forward: What’s next for AI at Tamr? Is there anything you are looking forward to?

At Tamr, we're constantly pushing the boundaries of AI in data management. We understand that the volume, velocity, and variety of data are constantly evolving in this AI-driven era, and business needs are changing right along with it. 

That's why we're focused on continuous innovation. We're excited to explore new frontiers in AI that will further automate data mastering tasks, improve data quality even more, and provide even greater flexibility for handling diverse and dynamic data sets. 

Ultimately, our goal is to stay ahead of the market and deliver solutions that empower businesses to fully unlock the potential of their data, no matter what the future holds.

******

Make sure you check out the latest edition of AI Magazine and also sign up to our global conference series - Tech & AI LIVE 2024

******

AI Magazine is a BizClik brand 

Share

Featured Articles

AWS Bedrock Gets Anthropic's New Claude 3.5 Sonnet Model

Amazon Bedrock and Anthropic upgrade their partnership to build Gen AI applications... with Gen AI assistance by adding Claude 3.5 Sonnet to copilot tasks

What Dell and Super Micro can Bring Musk’s xAI Supercomputer

Elon Musk's xAI partnership with server hosting titans Dell and Super Micro could see his ambition for 'the world's largest supercomputer' lift off

Toshiba Takes Another Step to Ushering in Embodied AI

Toshiba's Cambridge Research Lab has announced two breakthroughs in Embodied AI alongside a new group to renew focus on the tech

Why AWS is Investing $230m in Credits for Gen AI Startups

Cloud & Infrastructure

How Retrieval Augmented Generation (RAG) Enhances Gen AI

AI Applications

Synechron’s Prag Jaodekar on the UK's AI Regulation Journey

AI Strategy