Jun 2024 – Aug 2024 · 2 mos · Berkeley, California, United States

• The ML Alignment & Theory Scholars (MATS) Program is an independent research and educational seminar program that connects talented scholars with top mentors in the fields of AI alignment, interpretability, and governance. I will conduct research while also attending talks, workshops, and networking events with other members of the Berkeley alignment research community. The MATS program is an initiative supported by the Berkeley Existential Risk Initiative and historically by Open Philanthropy.

Oct 2023 – Apr 2024 · 6 mos · Kraków Old Town, Małopolskie, Poland

• To promote human resource capacities in the developing countries and to enhancing international understanding and technical collaboration among nations, UNESCO Chair for Science, Technology and Engineering offered thirty (30) fellowships to researchers across the globe.

Jul 2023 – May 2024 · 10 mos · Fairfax County, Virginia, United States

• Learning the key ideas in Austrian, Virginia, and Bloomington schools of Political Economy and learning how to utilize these ideas in academic and policy research. These three schools combined help create a comprehensive framework to analyzing the social organization, coordination, and prosperity. The Austrian school helps improve understanding of the capabilities of decentralized, emergent orders as well as the limits of centralized monocentric approaches to governing. The Virginia school of thought uses the economic way of thinking to study politics realistically. And the Bloomington school examines the effectiveness of polycentric systems for self-governing systems.
• My track in the fellowship was "Governance." Here, I study and introspect on National Economic Planning, Socialism, Marxism, and Economic Coordination, Community Revival in the Wake of Disaster, role of Local businesses and public-private partnerships at the time of crisis, disaster, governance, and entrepreneurship, Humanitarian Action and the "Planner's Problem", how humanitarian aids cripples self-sufficiency of a nation and viciously alters the economic DNA of a nation and compromise sovereignty, Classical Liberalism, Mainline Economics, and Social Movements, etc.
• I utilized these learnings to understand and outline the political economy of AI that navigates the nexus of politics, economics, and technology. Governments grapple with regulation and societal impact, while balancing AI's potential for growth with concerns of job displacement and inequality. Economically, AI promises productivity gains but raises questions about market competition and wealth distribution. Ethical considerations, such as bias in algorithms and geopolitical tensions, further shape this landscape. I believe, understanding these dynamics is crucial for crafting policies that harness AI's potential while addressing its societal challenges.

Jan 2021 – Nov 2022 · 1 yr 10 mos · Bengaluru, Karnataka, India

• Worked primarily with the Data Insights and Core Engineering team at Google Cloud PA with a focus on developing and maintaining the ETL SQL pipelines and improving the robustness and strong test coverage and designing test efficacy metrics.

May 2020 – Jul 2020 · 2 mos

• Worked with Mozilla's Fix-The-Internet Spring Lab. Streamlined tasks, user journey, critical features, built prototype for an app to help detect and explain cleanliness levels of the surface.

Jan 2020 – Jan 2021 · 1 yr · Delhi, India · On-site

• My Master's thesis focused on designing therapeutic treatments for Neurological Disorders such as Dementia using Neural Correlates of Motion Processing and the concept of Attentional Blink, done under the supervision of Dr. Sonia Baloni Ray in her Signal Processing and Visual Cognition Lab.
• Collected, pre-processed, and done signal processing and analysis on continuous EEG data to understand how attention works in human beings when the stimuli is in motion and to detect the presence of attentional blink in an RSVP task.
• Toolboxes: MATLAB, EEGLAB, Chronux, Brain Vision Analyser, FeildTrip.
• Lab: https://sites.google.com/iiitd.ac.in/sonia-baloni-ray/home?authuser=0

Jan 2020 – Jul 2020 · 6 mos · Noida, Uttar Pradesh, India · On-site

• Worked primarily on the iOS application features for the Photoshop App (Creative Cloud) at Adobe.

Sep 2019 – Dec 2019 · 3 mos · Delhi, India · On-site

• Project: Bware - A safety compass and Imperative for Travellers in Delhi
• Media Coverage in The Indian Express: https://www.newindianexpress.com/cities/delhi/2019/Dec/01/bware-four-young-students-come-up-with-app-that-aims-to-make-travelling-in-delhi-safer-2069569.html
• Built a real-time heat map, integrated with Google Maps, to provide Delhi travellers a way to navigate safely and be well-informed about the possible annoyances along their route (crowdsourced). Designed and led experiments for leveraging the human-centered design process to collect data and define clear user requirements for feature prototypes.

May 2018 – Sep 2018 · 4 mos · Delhi, India · On-site

• Title: Medical Assistant Hindi Language Chatbot for Breastfeeding Mothers
• Built an interactive chatbot for breastfeeding mothers in the Hindi language. Collected, extrapolated, and scraped over a 1000 Q/As in Hindi Language. Retrieval-based model used. Proposed a heuristic for best response based on the input and context.
• Research Report: https://drive.google.com/file/d/1aN0jVzX0KUYehlKpU8fPE7aoSb80TUXa/view

Jan 2018 – Apr 2018 · 3 mos · Delhi, India · On-site

• Artistic Style Transfer
• Achieved artistic style transfer using image processing paradigm, instead of resource-consuming deep neural nets. Accomplished “style synthesis” from the input art styles and fused it with the input images.
• Tools: MATLAB

Sep 2019 – Dec 2019 · 3 mos · California, United States

• Worked with PIs from Stanford's Public Health Care and Social Good Dept. Researched and Designed novel metrics for Ethics and Fairness in Gender Based Violence, analysed downfalls of Risk Assessment Instruments in legal system; Retrospective and Qualitative analysis of a women’s helpline using AI-NLP; Contextual Inquires with helpline users to understand choices and needs.
• During this work, I also conducted thought experiments, gathering a woman's perspective on safety and privacy, and helped better understand the parameters which, in hindsight, drive your decision making in a social context. This was a study I performed to better understand the Ethics and Fairness in Gender Based Violence. Glimpse: https://sites.google.com/iiitd.ac.in/safetycompass/home

Jul 2018 – Jun 2021 · 2 yrs 11 mos · Delhi, India · Hybrid

• RA at the Multilingual Technologies and Computational Science Lab under Prof. Tayal and Dr. Jain.
• I adopted user-centered approaches to solving sensitive problems within healthcare and national defense contexts.

Jul 2018 – Sep 2018 · 2 mos · Remote · Remote

• Mentor: Louis Jenkins, PhD, High Performance Computing, University of Rochester
• Project: Chapel Parallel Programming Language
• Summary: Learnt design and analysis of algorithms, object-oriented prototyping, and parallelism. Implemented sorting algorithms in Chapel and studied their performance and comparisons with existing implementation.
• Blog: https://railsgirlssummerofcode.org/blog/2018-10-29-team-sectumsempra