Bhiman Kumar Baghel

I'm

My research interests are
Natural Language Processing Conversational AI NLU Common sense Reasoning AI Fairness

About

I am 1st Year Computer Science PhD Student at the University of Pittsburgh, PA, USA. In my PhD, I'm eager to tackle key challenges in grounding language models for commonsense reasoning, improving their fairness and reliability for everyday use. My research background lies in Natural Language Understanding (NLU) for Conversational AI, including multi-intent classification and managing out-of-domain queries. A significant milestone in my career was leading the overhaul of the Samsung SmartThings machine learning architecture. During this, I worked as Lead Engineer - NLP at Samsung Research Institute, Bangalore, India. I devised solutions for NLU research challenges faced by Samsung's Conversational AI - Bixby, from research to commercialization. I post-graduated from the Indian Institute of Technology (IIT), Kharagpur as an MTech in Computer Science and Engineering in 2019. Before this, I received my BTech in Computer Science and Engineering from Dr B R Ambedkar National Institute of Technology (NIT), Jalandhar in 2017.

Publications

Patents

News

Feb'24
A1 US Patent published - Method and system for mitigating physical risks in an IOT environment

Feb'24
Attended Google Research Week

Jan'24
Started working as Teaching Assistant - Intro to Natural Language Processing @ SCI, UPitt

Dec'23
Sucessfully Completed 1st Semeter of CS PhD @ SCI, UPitt

Sep'23
Started working as Teaching Assistant - Operating System @ SCI, UPitt

Aug'23
Started CS PhD @ University of Pittsburgh, PA, USA

Mar'23
Promoted to Lead Engineer - NLP @ SRIB

Mar'23
Recived Samsung Excellence Award @ SRIB

Mar'23
A1 US Patent published - METHODS AND SYSTEMS FOR DETERMINING MISSING SLOTS ASSOCIATED WITH A VOICE COMMAND FOR AN ADVANCED VOICE INTERACTION

Mar'23
Recived MBO High Perfromance Bonus @ SRIB

Sep'22-Dec'22
Business trip to Samsung HQ, South Korea

Publications & Patents

  • All
  • Conversational AI
  • Patents
  • 2023
  • 2021

Resume  

Education

Doctor of Philosophy (PhD) Computer Science

2023 - Present
University of Pittsburgh, PA, USA

Courses Taken

  • Advanced Topics in NLP (Seminar)
  • Intro to NLP
  • Distributed Operating System

Master of Technology (M.Tech) Computer Science & Engineering

2017 - 2019
Indian Institute of Technology (IIT), Kharagpur, India

Bachelor of Technology (B.Tech) Computer Science & Engineering

2013 - 2017
National Institute of Technology (NIT), Jalandhar, India

Academic Experience

Teaching Assistant - Intro to Operating Systems

Sep'23 - Dec'23
University of Pittsburgh, PA, USA

  • Conducted weekly recitations for class of ~50 undergrad students. Graded quizes and assignments.

Teaching Assistant - Programming and Data Structure & DBMS

Jul'18 - May'19
Indian Institute of Technology (IIT), Jalandhar, India

  • Thought lectures on Programming & DBMS concepts to a class of 80+ undergrad students. Evaluated lab assignments and tests based on those concepts.

Professional Experience

Lead Engineer - NLP

Jun'19 - Aug'23
Samsung Research Institute, Bangalore, India

  • Role: Research, Design and Develop solutions for Bixby’s (Conversational AI) industry‑level NLU challenges.
  • CoSMIC ‑ Contextual SmartThings Multi‑Intent Composer: Researched and commercialized BERT‑based Multi‑Intent NLU for SmartThings. Achieved 96% E2E Accuracy with 12% performance boost. Reduced the error rate by 67%.
  • Conversational AI: Worked on complex scenarios of Multi‑Intent and Slot Classification, Zero-shot Out‑Of‑Domain Detection, Intelligent Device Resolution service, Bixby Recall ‑ Chat Question Answering (QA) System and Knowledge Representation.
  • Achievements: Publications (x2), Patents (x4), MBO High Performance Bonus (x3), Samsung Excellence and other awards (x5), CLab Finalist

Extreme Blue Intern

May'18 - Jul'19
IBM, Bangalore, India

  • Project: Template based Machine Learning Models for MVS
  • Designed and developed a time series forecasting, self‑learning, DL system to predict, prevent or auto‑resolve or assist in IT infrastructure Failure.
  • Achievement: 2nd Runner Up ‑ Audience Poll Category, IBM Extreme Blue Expo