Research Engineer / Realist
I work on Large Language Models at Zoho.
Prev, I used to work on using RL for DB Autotuning and Query Optimisation. Read about it **here.**
Currently, my research interests focus on building trustworthy AI through the intersection of:
→ statistical post-training methods for alignment and uncertainty quantification,
→ neuro-symbolic systems for faithful reasoning, verification, and interpretability.
I am also studying the different ways we can look at the impact of data distribution on large model training.
You can reach out to me **@sauravlmx or drop an email at : [email protected]**
I consider myself a hardcore realist to the point that I have come up with a Realist’s Manifesto.
I maintain a list of my favorite blogs on systems, machine learning and meta engineering. Access it **here.** There is also a reading list (of books) I have partially read, or I want to read. Access the reading list **here.**
I’m no writer by trade; I write to make sense of my thoughts. More about it here
When I learn something new, I feel compelled to share about it. I write not only for clarity but also to channel the excitement of learning new things into a valuable resources for others. Feel free to give me feedback.