Sarang Kulkarni on Lessons from Building Deep Research Agents in Production

Deep Research Agentic systems are AI agents that perform multi-step internet research for complex tasks using dynamic reasoning, multi-hop information retrieval, and structured analytical report generation. In healthcare and clinical trials, they support needs beyond simple question answering by discovering, connecting, and reasoning across internal and internet data while maintaining reliability, transparency, and compliance. Drug development is costly and many studies proceed without prior evidence due to broken access to existing knowledge. A RAG-based chatbot initially handled simple study queries, but complex questions required an agentic RAG approach. For deep research, an Agentic RAG++ system was built with clarification loops, iterative retrieval, and agentic loops to improve outcomes.

"Deep Research Agentic Systems, such as OpenAI and Gemini Deep Research Agent, are AI Agents designed to conduct multi-step research on the internet for complex tasks using dynamic reasoning, multi-hop information retrieval, and generate comprehensive, structured analytical reports at the level of a research analyst."

"In critical industries like healthcare and clinical trials, the researchers need more than the traditional AI models that perform simple Q&A tasks. They need systems that can discover, connect, and reason across both internal and Internet data, while maintaining reliability, transparency, and compliance."

"Kulkarni started the presentation by highlighting that it typically costs $2.6B to bring a new drug to market. Also, about half the research studies are conducted without prior evidence because the knowledge exists, but access to this knowledge and information is broken. In the overall drug discovery and development pipeline, getting the right data at the right time is a major challenge."

"For simple queries in the study, the RAG solution worked fine, but for complex questions, they had to enhance it to be an agentic RAG [] application. And for deep research use cases, the team developed a solution they call the Agentic RAG++."

Read at InfoQ