Sagnik Anupam

Jan 2022 - Present

I am currently developing neurosymbolic machine-learning models using PyTorch that can use reasoning to solve problems in mathematical and computational domains under Dr Omar Costilla-Reyes and Dr Armando Solar-Lezama in the Computer-Aided Programming Research Group at MIT CSAIL. In the 2022-23 academic year, my work was funded under the MIT SuperUROP program, where I was selected as an MIT EECS Gerstle Research and Innovation Scholar (2022-23).

May 2022 - Aug 2022

I worked as an AI/ML engineering intern at FindOurView, a text-response and consumer-feedback analysis startup. There I built transformer-based Natural Language Processing (NLP) models for various text classification and generation tasks, which reduced manual effort in data inspection by 70% and sped up execution by 15% in key steps of the analysis pipeline. I improved system performance by designing better dataset preprocessing methods and combining a wide variety of models, reaching an overall accuracy of >90% on some natural language processing tasks.

Sep 2020 - May 2022

I worked as an undergraduate researcher at the MIT Research Laboratory of Electronics (RLE) in the Speech Communication Group under Dr. Stefanie Shattuck-Hufnagel and Dr. Jeung-Yoon Elizabeth Choi on acoustic cue detection.

There, I developed tools using C and MATLAB to detect and interpret acoustic cues to distinguish among speech sounds using sampled audio. I also built a dataset of 350 labeled spectrograms and built models to identify glottal-tier acoustic cues, achieving an accuracy of 77% (17% more than current best probabilistic models).

I presented the results of my research at the Speech Communication, Psychological and Physiological Acoustics, and Signal Processing in Acoustics poster session at the 181^st meeting of the Acoustical Society of America in Seattle, WA.

Jan 2021 - May 2021

I worked with the MIT Sandbox Innovation Fund to analyze data from ~ 1600 teams that obtained seed funding for their venture from the MIT Sandbox program over 4 years.

Jul 2019 - May 2020

I was a software development intern at SincGrid, where I worked on a project to integrate voice assistants with IoT devices for automating sensor data collection for an embedded-systems startup. I built the cloud-based software using Python, Alexa Skills Kit, and Google Dialogflow.

Feb 2023 - May 2023

Alongside my teammates in the AIM Labs Spring 2023 cohort, I worked on Briefly, a document summarizer built using industry-standard GPT-4 capabilities. Our software used contextual information about a reader's background (such as their age, domain of expertise, profession, etc.) to build a customized summary of a document easily understandably by someone of their background.

Feb 2023 - May 2023

Alongside my teammate in the MIT class 6.S986: Large Language Models and Beyond (taught during the Spring 2023 semester by Prof. Yoon Kim) I worked on CapText, a caption-generation technique relying on using only image description and image context for captioning models, which beat state-of-the-art image captioning models on the CIDEr metric.

Sep 2022 - Dec 2022

Alongside my teammates in the MIT class 6.5210: Advanced Algorithms (taught during the Fall 2022 semester by Prof. David Karger), I designed a 1/3-approximation algorithm for the Multiple Drone-delivery Scheduling Problem (MDSP), as well as a dynamic-programming algorithm to exactly solve the unit-cost variant of the problem.

Aug 2018 - Nov 2019

Under Dr. Padmini Pani of Jawaharlal Nehru University (JNU), I built Augura, a flood-forecasting model for the Brahmani-Baitarani river basin in Odisha, India. For building the machine-learning model, I utilized Python, NumPy, scikit-learn, extreme learning machines, and particle swarm optimization.

Citation: Anupam, S., & Pani, P. (2020). Flood forecasting using a hybrid extreme learning machine-particle swarm optimization algorithm (ELM-PSO) model. Modeling Earth Systems and Environment, 6(1), 341-347. https://doi.org/10.1007/s40808-019-00682-z.

Aug 2017 - Feb 2019

Under Dr. Arpan Kar of Indian Institute of Technology (IIT) Delhi, I conducted a comparative study to identify optimal machine-learning models for phishing website detection, achieving 90.3% accuracy. For developing this project, I used Python, NumPy, EvoloPy, scikit-learn, and nature-inspired optimization algorithms.

Citation: Anupam, S., & Kar, A. K. (2020). Phishing website detection using support vector machines and nature-inspired optimization algorithms. Telecommunication Systems, 1-16. https://doi.org/10.1007/s11235-020-00739-w.

May 2021 - Present

I am a brother of the Alpha Delta Phi fraternity, where I served as Vice-President (Fall 2023), Housing Manager (2022-23), Social Chair (Fall 2022), Literary Chair (Fall 2021), and Summer House Manager (Summer 2022).

Jan 2023 - Present

I am a member of the MIT Assassin's Guild, MIT's live-action roleplaying society. Games I have participated in include: HP7 (IAP 2023), Persephone's Gift (Spring 2023), and An Aquarium Auction (Spring 2023).

Feb 2023 - May 2023

I was a member of the MIT AI Alignment (MAIA) AI Safety Fundamentals program, as a part of their Spring 2023 cohort.

Feb 2023 - May 2023

I was a member of the AIM Labs Spring 2023 Cohort. AIM Labs is an initiative to build public-facing machine-learning applications by AI @ MIT, a community of undergraduates aimed at promoting and fostering the growing interest around machine intelligence on campus.

Sep 2021 - May 2022

I was a participant in UPOP, a yearlong professional development program for MIT sophomores to build the relevant and critical skills necessary for a career in industry, research, or academia.

During the program, I completed multiple professional development workshops taught by MIT faculty and industry professionals, including a three-day Team Training Workshop.

Sep 2020 - May 2021

I was a member of the xFair and Dev Ops committees of TechX, a student-run organization dedicated to bringing cutting-edge technology and innovators to the MIT campus.

Sep 2020 - May 2021

I was a diligence associate at MIT Capital Partners (Spring 2021) and a sourcing analyst at Harvard Undergraduate Capital Partners (Fall 2020), both branches of Collegiate Capital Partners, a student-run early/growth-stage investment advisory firm.

Sep 2020 - May 2021

I served as a staff member and webmaster at Rune Magazine, MIT's art and literary magazine, for issue 42 (academic year 2020-21).

Sep 2020 - Aug 2021

I served as the MIT International Student Association's Class of 2024 representative during the academic year 2020-21, where I helped advocate for international freshmen. Towards the end of my freshman summer, I also helped serve as an International Orientation Mentor for incoming international students before the Fall 2021 semester.

Sep 2020 - June 2021

I also ran a blog-cum-newsletter called Curated, where I occasionally wrote about college life, literature, art, music, and languages. You can check out some past issues here. It is currently on hiatus, but will probably be brought back in another form sometime.

MIT EECS

In 2022, I was selected as an MIT EECS Gerstle Research and Innovation Scholar (2022-23) as a part of the MIT SuperUROP program.

CITADEL

In 2022, my team won the Citadel Prize for Highest EV of Bets on the Turn at MIT Pokerbots.

MIT CMSW

In 2022, I won the 1^st Prize in the Boit Manuscript Prize for Poetry and the 3^rd Prize in the Robert A. Boit Prize for Poetry, awarded as part of the Ilona Karmel Writing Prizes by MIT Comparative Media Studies/Writing.

In 2021, I won the 2nd Prize in the Ellen King Prize for First-Year Writing, awarded as part of the Ilona Karmel Writing Prizes by MIT Comparative Media Studies/Writing.

INTEL ISEF

I represented India at the Intel International Science and Engineering Fair (ISEF) in 2018 (at Pittsburgh, PA) and 2019 (at Phoenix, AZ) after winning Grand Awards at the IRIS National Science Fair.

In 2018, I participated in the Robotics and Intelligent Machines category with my project "Phishing Website Detection Using Support Vector Machines and Nature-Inspired Optimization Algorithms", and in 2019 I participated in the Earth and Environmental Sciences category with my project "Augura: Flood Risk Prediction Using Machine-Learning and Geographic Information Systems."

At ISEF 2019, my project won the Third Grand Award in the Earth and Environmental Sciences Category, as well as a Second Award from the Ashtavadhani Vidwan Ambati Subbaraya Chetty (AVASC) Foundation.

INTERNATIONAL LINGUISTICS OLYMPIAD

I won an Honorable Mention at the International Linguistics Olympiad (IOL) in 2017 at Dublin, Ireland, after obtaining a Silver Medal at the Panini Linguistics Olympiad.

ASIA-PACIFIC LINGUISTICS OLYMPIAD

I won a Bronze Medal at the Asia-Pacific Linguistics Olympiad (APLO) in 2019.

INTERNATIONAL PHILOSOPHY OLYMPIAD

I won a Bronze Medal at the International Philosophy Olympiad (IPO) in 2018 at Bar, Montenegro.

INTERNATIONAL YOUTH PHYSICISTS' TOURNAMENT

I was a part of the team that represented India at the International Youth Physicists' Tournament (IYPT) in 2018 at Beijing, China.

GOOGLE

I was selected to be a member of Google's CSSI-Coursera 2020 cohort.

GOOGLE INDIA

I won the Google India Code to Learn Contest in 2019 in the machine learning category.

MINISTRY OF WOMEN AND CHILD DEVELOPMENT, GOVT. OF INDIA

I was awarded the National Child Award for Exceptional Achievement (or Pradhan Mantri Rashtriya Bal Puraskar), India's youth civilian honor, in 2020 in the innovation category.

NEW YORK ACADEMY OF SCIENCES

I was a member of the Junior Academy of the New York Academy of Sciences (NYAS) from 2019 to 2020.

NASA AMES

I won a First Award in the Grade 11 Large Team category at the NASA Ames Space Settlement Design Contest (now the NSS Space Settlement Contest) in 2018.

CONRAD SPIRIT OF INNOVATION CHALLENGE

I was a finalist in the Aerospace and Aviation category at the Conrad Spirit of Innovation Challenge in 2017 at the NASA Kennedy Space Center in Orlando, FL.

Jun 2019 - Jul 2019

I volunteered as a research intern in June 2019 as a part of a school-run program with the Sinhayana Foundation, an independent non-profit aiming to empower underprivileged children by providing dance education free of cost through their Dance Out of Poverty initiative. As a volunteer, I wrote a paper analyzing the dance sector and the developments in dance education across different countries.