Context.ai plans to combine product analytics and LLMs
Context.ai was launched this year to address the problem of assessing the performance of large language models (LLMs) companies use to improve customer interactions. They have recently raised approximately $3.5 million in seed investment to develop their solution further. GV (Google’s venture arm) and Theory Ventures co-led the investment.
Presently a small team with six employees, Context.ai focuses on future growth and emphasizes building a diverse and inclusive organization. Henry Scott-Green, Context.ai’s CEO, and Alex Gamble, CTO, highlighted a gap in the market. They highlighted that developers who create LLMs face consistent challenges in understanding user behavior and model performance. Most developers find their models like “black boxes,” lacking transparency.
Context.ai provides similar services to product analytics tools like Amplitude or Mixpanel. However, Context.ai analyses the data created by LLMs to check the quality and usefulness of their content in response to customer queries. Customers share chat transcripts with Context via an API that utilizes natural language processing (NLP) to analyze and categorize conversations by topic. It then analyses customer satisfaction with the responses, planning to enhance model effectiveness.
Context.ai highlights a prominent shift towards text-based interactions driven by LLMs. They identified the need for specialized tools in this new landscape. They initiated with a prototype, involving early customers and design partners, and have continually refined their product.
Context.ai focuses on prioritizing security and privacy. It plans to remove personally identifiable information (PII) during data ingestion. They do not use the content for model building or marketing purposes and retain it for a maximum of 180 days before deletion.
