Jigsaw: Large Language Models meet Program Synthesis
Used program synthesis techniques to generate code from multi-modal user input using large language models like GPT-3. Developed a Jupyter notebook extension to generate code from user commands and I/O examples for the Pandas library in Python. This project was done under the guidance of Dr. Sriram Rajamani at Microsoft Research, India.
Skanda Vaidyanath
Staff Research Scientist
My research interests lie primarily in the area of reinforcement learning (RL) and control to build agents that can acquire complex behaviours in the real world via interaction.
Talks
Jigsaw: Large Language Models meet Program Synthesis
Used program synthesis techniques to generate code from multi-modal user input using large language models like GPT-3