Skip to content
View joyjitchatterjee's full-sized avatar
😎
Coding
😎
Coding
16 followers · 8 following

Achievements

Achievement: Pull Sharkx2Achievement: StarstruckAchievement: Arctic Code Vault Contributor

Achievements

Achievement: Pull Sharkx2Achievement: StarstruckAchievement: Arctic Code Vault Contributor

Block or report joyjitchatterjee

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
joyjitchatterjee/README.md

Dr. Joyjit Chatterjee. Enterprise AI, product delivery and innovation.

Portfolio & impact · Career in detail · Publications · LinkedIn

I connect AI strategy with the work needed to deliver it: finding valuable problems, shaping products, building teams and bringing AI into everyday decisions.

Lead Data Scientist at EPAM Systems · Distinguished Visiting Fellow at Loughborough University · Visiting Academic at the University of Hull · PhD in Computer Science

My work spans enterprise AI consulting, generative and agentic AI, explainable decision support, manufacturing, R&D and the energy transition. I bring together technical delivery, domain expertise and senior stakeholders to move from discovery and proof of value to deployment and adoption.

From strategy to delivery

Enterprise AI AI products Operational impact
Contribute to multi-million-pound engagements at EPAM across Europe and North America, supporting Fortune 500 and FTSE 250 organisations. Led the AI workstream for a $200k Reckitt × Hull partnership, taking shelf-life prediction from research to a production application. The KTP received the highest grade, Outstanding. Contributed to approximately 30% waste reduction in major UK factories at Reckitt through data science and manufacturing analytics.

GenAI for procurement: Contributed to a negotiation intelligence platform for a global consumer goods company. Evaluation demonstrated savings potential equivalent to 2% of annual procurement spend. Project details

Client work, leadership responsibilities and evaluation results

At EPAM, my direct client delivery includes Reckitt and Syngenta Group. I also support pitches, pre-sales and use-case discovery for Qatar Airways, Unilever, Barry Callebaut, Brussels Airport Company, Cleveland Clinic Abu Dhabi, Adidas, Co-op, Currys and PPG Industries.

My responsibilities span discovery, RFP responses, proof of concept and value, project planning, governance, stakeholder alignment, mentoring and helping assemble multidisciplinary delivery teams.

At Reckitt, evaluations demonstrated potential annual savings of up to £300k in manufacturing and $780k plus 10,000 laboratory hours in R&D. These are evaluation-based potential benefits, not realised savings.

Explore the full responsibilities, projects and supporting detail →

Recognition & academic contribution

  • Forbes 30 Under 30 Europe, Manufacturing & Industry, 2022. Profile
  • British Data Awards 2026 finalist, Data Leader of the Year (Partner). Finalists
  • ETH Zurich, invited expert on a Doctoral Examination (Aptitude Colloquium) committee. Academic service
  • Green Talents 2021, one of 25 international awardees selected by Germany's Federal Ministry of Education and Research. Award profile
  • Horizon Europe's NGI Search, external evaluator and reviewer, 2023. Details

Research you can read, ideas you can hear

Latest paper: Graph-Grounded Multimodal Agent for Trustworthy Power-Line Operations and Maintenance
GLOW at ISWC 2026. Research combining visual evidence, knowledge graphs and physics simulation for traceable maintenance decisions.
Read or download the PDF · Publication library

Cover of Renewable Energy Transition with Artificial Intelligence

Renewable Energy Transition with Artificial Intelligence
Co-editor with Nina Dethlefs. A Wiley volume connecting research with energy-sector challenges.

Explore the book → 		Joyjit speaking on the Low to Grow podcast

Explainable AI and the Power of Quiet Thinking
On enterprise AI adoption, explainability and reflective leadership with Low to Grow.

Watch the conversation →

Selected open research

These repositories share resources accompanying my research in trustworthy industrial AI.

Repository What you can explore
XAI4Wind Domain knowledge graph and Neo4j resources for explainable wind turbine operations and maintenance.
WindTurbine-QAKG Question–query datasets, templates and graph resources for interactive maintenance decision support.
TurbineSCADACausality Supplementary temporal causal graphs and preprocessing material for research on wind turbine SCADA data.

Technical focus: Python · LLMs & RAG · Agentic AI · Azure OpenAI · Knowledge graphs & Neo4j · Explainable AI · Causal inference

Full portfolio · Google Scholar · ORCID · Writing · Bluesky · Email

Popular repositories Loading

  1. XAI4Wind XAI4Wind Public

    Supplementary data for utilising knowledge graph databases towards explainable decision support in wind turbines.

    Python 24 7

  2. WindTurbine-QAKG WindTurbine-QAKG Public

    Automated Question-Answering Over Knowledge Graphs in O&M of Wind Turbines

    Python 14

  3. DeepLearning-Tensorflow-Basics DeepLearning-Tensorflow-Basics Public

    This repository contains basic Python code for learning to use Tensorflow, along with its basic syntactic and operational intricacies, as a solid foundation for other complex Deep Learning projects.

    Jupyter Notebook 4

  4. PowerLine-KG-Corpus PowerLine-KG-Corpus Public

    Power-line maintenance KG and description datasets covering 255 InsPLAD images, with separate human and template provenance. GLOW @ ISWC 2026.

    Cypher 4

  5. Vedic-Binary-Multiplier Vedic-Binary-Multiplier Public

    The project involved design of a Binary Multiplier which is highly efficient in terms of computation time and used the centuries old Urdhava-Tiryakbham algorithm. Structural Coding was used to firs…

    VHDL 2

  6. TurbineSCADACausality TurbineSCADACausality Public

    Supplementary material for our TORQUE 2020 Submission "Temporal Causal Inference in Wind Turbine SCADA Data Using Deep Learning for Explainable AI"

    2