Athresh Guruprakash

Athresh Guruprakash: Advancing Responsible AI Through Leadership, Research, and Education

With more than 18 years of experience in software engineering, technology leadership, and applied artificial intelligence, Athresh Guruprakash has built a career defined by sustained achievement and increasing professional responsibility.

As a Senior Software Engineering Manager at Equifax, Guruprakash leads engineering teams working within financial-services risk technology—an environment where technical innovation must coexist with regulatory compliance, operational resilience, and responsible decision-making. His work requires an understanding of regulatory frameworks such as the Fair Credit Reporting Act, the Equal Credit Opportunity Act, Basel III, and SR 11-7.

His responsibilities extend beyond delivering technology solutions. Guruprakash has built and managed diverse, cross-functional teams while mentoring professionals at multiple career stages, from early-career developers to senior individual contributors. His leadership reflects a long-term commitment to developing both dependable systems and capable engineering teams.

Guruprakash has also contributed to the profession through original research addressing some of the most pressing issues arising from the adoption of artificial intelligence.

His paper, “Confidently Wrong: The Hidden Cost of LLM Hallucination in Credit Decisioning,” examines the risks created when large language models generate inaccurate or unsupported information in credit-risk and lending environments. Rather than treating hallucination as a purely technical problem, the research evaluates its operational and business consequences.

The paper introduces practical concepts, including the Hallucination Productivity Tax model, which is intended to quantify the operational cost of errors generated by large language models. It also presents a dataset-tuning blueprint for developing credit-specific language models and introduces the concept of Deployment Half-Life to help organizations evaluate how the reliability of an AI system may change after deployment.

His broader research covers applied machine-learning methodologies, including the combination of structured machine learning with retrieval-augmented context for outcome prediction. He has also examined the architectural challenges involved in designing live-video-streaming systems capable of withstanding sudden increases in demand at a massive scale.

In another paper, “From Scripted Automation to Intelligent Quality Engineering,” Guruprakash traces the development of software testing from traditional automation tools such as Selenium to emerging AI-assisted quality-engineering practices. The work draws upon his practical experience in testing and engineering leadership while exploring the direction in which the industry is progressing.

Taken together, these contributions demonstrate an ability to connect technical research with real-world implementation. His work does not stop at identifying complex problems; it proposes frameworks that engineers, technology leaders, and regulated organizations can apply in practice.

Guruprakash’s contribution also extends beyond corporate and research environments. Through Tech with AG, an educational initiative, he works to make complex artificial-intelligence and machine-learning concepts accessible to students, early-career developers, and non-technical learners.

The initiative presents topics such as retrieval-augmented generation, vector databases, agentic frameworks, and applied AI-system design in a structured and understandable manner. This educational work reflects his belief that professional achievement carries a responsibility to share knowledge and help prepare the next generation of technology practitioners.

Guruprakash’s career brings together three qualities central to Fellowship recognition: sustained technical leadership, original contribution to professional knowledge, and service to the wider technology community. His record demonstrates not only achievement within his own roles but also a broader commitment to responsible innovation, professional development, and the advancement of computing and engineering.