Manav Nitin Kapadnis

Associate Consultant

United States3 yrs 9 mos experience

Highly Stable

Key Highlights

Expert in developing LLM-based applications.
Proven track record in NLP and machine learning projects.
Strong background in recommender systems and data analytics.

Stackforce AI infers this person is a Machine Learning and NLP specialist with a focus on healthcare and tech industries.

Contact

Skills

Core Skills

Large Language Models (llm)Machine LearningNatural Language Processing (nlp)Deep LearningData AnalyticsConversational Language UnderstandingRecommender Systems

Other Skills

Amazon Web Services (AWS)Amazon Web Services (AWS) SageMakerAnalyticsArduinoCC++Computer VisionData AnalysisKerasKnowledge GraphsMatplotlibNumPyOpenCVPandasPredictive Modeling

About

TLDR: Master's at CMU LTI Bachelor's at IIT Kharagpur Professional Background: Internships at Apple ML Foundations team, Sony Research R&D division, Triomics, and Inria Research Interests: Applied ML projects, LLM Agents, Knowledge graph-based recommendations, LLM evaluations for conversational understanding and path planning, and efficient RL training techniques on LLMs. Reachout email: iammanavk@gmail.com / mkapadni@andrew.cmu.edu

Experience

Carnegie mellon university - school of computer science - language technologies institute

2 roles

Teaching Assistant : NLP

Aug 2025 – Present · 7 mos · Pittsburgh, Pennsylvania, United States · On-site

Teaching Assistant - NLP

Jan 2025 – May 2025 · 4 mos · Pittsburgh, Pennsylvania, United States · On-site

Apple

ML Research Intern

May 2025 – Aug 2025 · 3 mos · Cupertino, California, United States · On-site

Proactive Intelligence Team
TLDR -> 1. LLM Agents for UI navigation
2. Semantic Search and retrieval for Shortcuts app.

Large Language Models (LLM)Machine Learning

Carnegie mellon university

Teaching Assistant - NLP

Aug 2024 – Dec 2024 · 4 mos · Pittsburgh, Pennsylvania, United States

Triomics

Research Engineering Intern

Jan 2024 – Apr 2024 · 3 mos · Remote

Engineered a state-of-the-art OncoLLM model using Qwen2 family of LLMs with expertly curated, human-annotated clinical trials data
Coordinated with annotators and nurses in San Francisco hospitals to curate high-quality oncology datasets for training and evaluation.
Implemented a comprehensive comparative analysis of performance across various SoTA LLMs and fine-grained oncology NER tasks.

Natural Language Processing (NLP)Deep LearningPyTorch

Awl, inc.

AI R&D intern

Jun 2023 – Aug 2023 · 2 mos · Sapporo, Hokkaido, Japan · Remote

Data AnalyticsDeep Learning

Inria

Research Intern @ Articulabo, COML

May 2023 – Jun 2023 · 1 mo · Paris, Île-de-France, France · On-site

Devised a comprehensive conversational grounding dataset to evaluate across state of the art LLMs using diverse perplexity metrics
Achieved 100% accuracy in GPT-4’s perplexity tests for Repair and 80% for Reference Ambiguity, compared with GPT-3.5 and Llama.
Concluded that GPT-4’s high conversation grounding performance ratios suggests a promising direction for future dialog system research.

Large Language Models (LLM)Conversational Language UnderstandingData AnalysisNatural Language Processing (NLP)

Sony research india

Data Science Intern

Dec 2022 – May 2023 · 5 mos · Bangalore Urban, Karnataka, India · Remote

Project Name: Attention-based Contrastive Learning for Knowledge-aware Recommender Systems
Achieved 8.77% increase in NDCG@20 by developing a knowledge-aware recommendation system using graph attention networks
Improved existing models with noise injection strategy achieving 0.279 NDCG@20 score and 0.4156 MRR@20 score on ML-100K dataset
Optimized using a noise injection strategy, and achieving 2.2% rise in Recall@20 on ML-100K and 11.13% leap in MRR@20 on ML-1M

Recommender SystemsKnowledge GraphsAmazon Web Services (AWS)