Dr. Kousik Naskar

AI Researcher

Kolkata, West Bengal, India8 yrs 8 mos experience

AI EnabledHighly Stable

Key Highlights

Expert in data-driven analytic solutions for scientific problems.
Proficient in advanced machine learning and deep learning techniques.
Strong problem-solving skills in collaborative environments.

Stackforce AI infers this person is a Research Scientist specializing in Data Science and Machine Learning for scientific applications.

Contact

Skills

Core Skills

Data ScienceMachine LearningArtificial Intelligence (ai)Statistical Data Analysis

Other Skills

ARIMAXAWS SageMakerAmazon Web Services (AWS)Analytical SkillsAnalytical SolutionsAnalyticsArtificial Neural NetworksC++CERN ROOTCNNCommunicationComputational PhysicsConvolutional Neural Networks (CNN)Data CleaningData Interpretation

About

Research professional with experience in data analytics and science, specializing in building state-of-the-art data-driven analytic solutions for various scientific and physics problems. Proficient in statistical data analysis, machine learning, deep learning, NLP, large language models, and generative AI, Agentic AI, along with technologies such as Python, C++, SQL, NoSQL, and Vector databases. Strong problem-solving skills with the ability to work in both independent and collaborative environments to investigate, develop, and test innovative solutions.

Experience

8 yrs 8 mos

Total Experience

5 yrs 9 mos

Average Tenure

Current Experience

Cern

2 roles

Member of CMS collaboration at CERN

Sep 2022 – Sep 2024 · 2 yrs

I continued my postdoctoral position at Iop, Bhubaneswar, in collaboration with CERN.

Member of CMS collaboration at CERN

Dec 2015 – Aug 2022 · 6 yrs 8 mos

As a valued member of the CMS collaboration during my PhD at IIT Bombay, I contributed to analyzing the proton substructure by examining the properties of particles generated in heavy-ion collisions. Analyzed petabytes (PB) of data to uncover rare physics processes. Additionally, I have done Monte Carlo event generation, detector simulation, and estimated relevant parameters using a statistical approach. These efforts contribute significantly to advancing future experiments in pursuing fundamental physics.

Iop

Postdoctoral Researcher

Sep 2022 – Sep 2024 · 2 yrs · India · On-site

Contributed to novel research in lepto-quark searches, enhancing tau particle
identification through advanced statistical, machine learning and deep learning techniques.
Implemented the DeepTau (CNN-based deep learning algorithm) discriminator to reduce tau misidentification w.r.t jet, electron, and muon. It reduces jet misidentification by over 50% and electrons by up to 90%, significantly improving detection accuracy.
Developed an interactive Multimodal Conversational Assistant application using open-source technologies, enabling users to upload documents for effective information retrieval through text, table, and image extraction with RAG techniques.
Leveraged document embeddings, LangChain framework, Google GEN-AI, and prompt engineering to enhance real-time response generation in an intuitive interface, showcasing expertise in multimodal response generation and UX design.

statistical data analysismachine learningdeep learningNLPPythonC+++5