Inside iMerit: How Radha Basu Built a Global AI Data Powerhouse with a Human Touch

Radha Basu’s Journey from HP to iMerit

Radha Basu, the Founder and CEO of iMerit, spent two decades at HP where she rose to lead the Enterprise Solutions group. Following that, she served as CEO of Support.com, taking the company public. In 2007, Radha co-founded the Anudip Foundation with Dipak Basu, and later launched iMerit in 2012 with a mission to empower marginalized youth in Bhutan, India, and New Orleans through meaningful tech careers.

The Vision and Mission Behind iMerit

iMerit specializes in delivering multimodal AI data solutions by integrating automation, expert human annotation, and advanced analytics. Their goal is to provide high-quality data labeling and model fine-tuning at scale, supporting critical sectors such as autonomous vehicles, medical AI, and technology.

Radha observed that many talented young people in underserved areas lacked opportunities despite India’s tech boom. By showing them how software powers advanced AI technologies, iMerit grew with a diverse team, notably with half the workforce being women, emphasizing adaptability and continuous learning.

Growth and Evolution of iMerit

Starting from a small team, iMerit has expanded into a global leader working with over 200 clients including eBay and Johnson & Johnson. The company evolved from a data annotation provider to a full-stack AI data service, covering annotation, validation, auditing, and red-teaming within the human-in-the-loop (HITL) lifecycle.

Their expertise spans managing perception stacks for autonomous mobility, clinical imaging AI in healthcare, and GenAI tuning in high-tech sectors. iMerit’s success stems from cultivating cognitive expertise that challenges and coaches AI models, and maintaining strong long-term client partnerships.

Technological and Strategic Milestones

Key milestones include securing their first $1M MRR deal in autonomous vehicles, rapid adaptation during the COVID-19 pandemic by shifting to fully remote operations, and acquiring Ango.ai in 2023. This acquisition enhanced their AI data labeling and workflow automation platform, expanding capabilities in radiology, sensor fusion, and GenAI fine-tuning.

Addressing Challenges in Critical AI Domains

Data quality is paramount in mission-critical sectors. iMerit focuses on avoiding severe errors by leveraging human insight to handle exceptions and edge cases. Their security measures comply with standards like SOC2, HIPAA, and TISAX to protect sensitive data.

Customization is key; their engineers build tailored integrations and reports to meet unique customer requirements, rejecting a one-size-fits-all approach.

Combining Robotics with Human Intelligence

Radha advocates for Human-in-the-Loop (HITL) workflows that blend automation speed with human nuance and oversight. HITL safeguards accuracy by validating AI outputs, catching errors, and ensuring ethical standards, especially in sensitive or high-risk scenarios.

Examples include targeted annotation, complex reasoning for edge cases, and fine-tuning AI agents through Reinforcement Learning from Human Feedback (RLHF). Red teaming and scenario-based testing further enhance model robustness.

The Ango Hub Platform and Generative AI

Ango Hub integrates automation with HITL expertise, supporting critical phases like annotation, validation, and model fine-tuning. The platform facilitates human review of AI outputs to prevent hallucinations or logic errors and supports continuous learning to align AI with user goals.

The recently launched Ango Hub Deep Reasoning Lab (DRL) provides a unified interface for Generative AI tuning using chain-of-thought reasoning and real-time human feedback. This advanced platform leverages iMerit Scholars—subject matter experts who coach AI models in complex tasks such as advanced math problem solving.

Expert-in-the-Loop in Practice

Expert-in-the-Loop means integrating specialized human expertise to validate and improve AI outputs. For instance, an American tech company used iMerit’s network of healthcare professionals to evaluate and safely scale medical chatbot data annotation, achieving 99% consensus among nurse evaluators and significantly reducing project costs.

Maintaining Quality and Employee Growth

iMerit defines quality collaboratively with clients, emphasizing rapid validation by subject matter experts. With a 90% employee retention rate and the proprietary iMerit One training platform, the company invests heavily in learning, automation, and knowledge management to sustain excellence and career progression.

Advice for Aspiring AI Entrepreneurs

Radha recommends focusing beyond technology to deeply understand customer needs, risk tolerance, and long-term partnerships. Selecting aligned investors and board members is crucial for shared values, especially during challenges such as the pandemic. Building a profitable and inclusive company is key to lasting success in the tech industry.