AI Specialist

Date: Jun 6, 2024

Location: Chennai, TN, IN, 600004

Company: W. R. Grace & Co.

Job Description

AI specialist is part of the India Digital Operations Centre and will be responsible for developing AI/ML solutions for catalyst manufacturing (Operations, Maintenance, safety, Quality, Supply Chain, Logistics etc. that will drive improvement in Grace manufacturing capabilities across the world. Critical to success will also be requirements for the person to recommend ways to improve operations performance, maintenance performance, Production planning, Supply chain Metrics.


  • Understand business use of data and stakeholder requirements to support strategic business objectives.
  • Create and Develop pilot projects to demonstrate the value of the proposed AI solutions. Collaborate with development teams to iterate the Solutions towards the business value demonstration.
  • Collaborate with functional experts, data scientists, and project managers, to ensure successful delivery of AI solutions. Proactively illustrate the risks and mitigation plan to stake holders.
  • Develop data-driven decisions and insights using predictive models to support grace operations & monitor the performance of assets as a function of parametric adjustment for manufacture excellence.
  • Analyze large volumes of internal and external data using common data science tools (Python, SQL/NoSQL etc.) & various ML Models to deliver valuable insights for operational excellence.
  • Effectively summarize & communicate complex analytic results to a variety of audiences.
  • Be the central point of contact for technical clarifications for stake holder and customers. Develop solutions for the customer concerns and feedback.
  • Awareness & evaluation of latest technologies in AI field and evaluating them for manufacturing sectors.

Required Skills

  • Strong written and verbal communication skills
  • Strong interpersonal skills needed to manage and collaborate with multiple internal resources within the Grace organization and external resources, including management of vendor relationships.
  • Outstanding analytical skills including financial business case creation.
  • Ability to quickly learn details of Grace’s chemical processes and identify the impact of automation improvements in operations performance, maintenance performance, Production planning, Supply chain Metrics.

Required Qualifications

  • BS/MS in Engineering (Chemical) with minimum of 3+ years prior experience as AI/ML Specialist level role or BS/MS in any other back ground with minimum of 3+ years prior experience in Chemical Manufacturing
  • Knowledge and/or experience with data acquisition, preparation, and validation.
  • Good knowledge and experience in handling large volume data sets using Programming languages (python, SQL & other script languages) & various ML models to develop predictive models to make operational improvement decisions/business decisions.
  • Experience with cloud environments: data lake, data warehousing, Google Cloud preferred.
  • strong experience in advanced analytics, model building, statistical modeling
  • Good knowledge and experience in handling implementing ML models (such as regression models, Classification models, Clustering models, Deep Learning Models)
  • Proficient in below techniques is highly desirable.
  • Linear Logistic Regression, Decision Trees, Random Forests, K-Nearest Neighbours, Support Vector Machines.


Required Qualification - Cont..

  • Statistical techniques: ANOVA, Principal Component Analysis,
  • Deep Learning Methods: ANN, CNN, RNN, Transformers.
  • Good understanding and hands-on experience with time series and forecasting modelling (ARIMA, SARIMA, Holt winter etc).
  • Knowledge of Agile / Scrum is preferable.
  • Experience in deploying and monitor ML models in Operations, delivering data products to end-users. Experience with ML CI/CD pipelines.
  • Develop an optimized solution from predictive models & convert them into tool-based solution for operation/business team leaders.
  • Experience in Preventive maintenance predictive models, equipment reliability analysis models are preferable but not mandatory.