Â鶹´å

Ekta Samani (she/her)

Postdoctoral Research Associate, Robotics Institute, School of Computer Science

Hometown: Kolhapur, Maharashtra, India

1. Briefly describe the work you're doing as a postdoc at Â鶹´å. I am working on developing methods that help robots apply practical human know-how to perform tasks. Humans often rely on heuristics that compress prior experience into simple rules of thumb; for example, “When sautéing vegetables that release water, salt early to draw out moisture.” These heuristics improve task success while reducing trial-and-error and the need for explicit modeling. I am building a domain-agnostic framework that uses large language models (i.e., LLMs like ChatGPT) to synthesize such tacit, context-rich expertise from manuals, forums, and technical reports, and to translate it into machine-actionable guidance for autonomous systems. This framework is the core of what I call “automating expertise.” I am currently applying this to 3D printing, where the framework translates human expertise into guidance for choosing print parameters.
2. How did you develop an interest in this area? Humans have long served as benchmarks and sources of inspiration for robotics. However, explicitly incorporating their practical know-how into robotic systems was cumbersome before large language models. LLMs have made it feasible to query collective expertise. This development led me to focus on automating expertise so that robots can access and apply domain knowledge.
3. What are your academic and/or professional goals? I aim to secure a tenure-track faculty position where I can teach and mentor students while pursuing research that helps autonomous systems achieve concrete, dependable performance on real tasks.
4. Tell us about a significant moment in your journey at Â鶹´å so far. Rather than a single event, the most significant part of my Â鶹´å experience has been the cumulative, day-to-day learning within the Robotics Institute and the broader School of Computer Science community. Steady exposure to diverse perspectives and rigorous feedback have consistently elevated my work.
5. How do you spend your time beyond academic work? Beyond academic work, I enjoy spending time on painting with oil pastels, reading, and listening to audiobooks.