Shashank Srivastava

Worked on end-to-end Marketing Mix Modeling (MMM) projects for multiple clients, analyzing large datasets to identify key drivers of business performance. Delivered actionable insights on media effectiveness and budget optimization within tight timelines. Successfully transitioned into this new domain, ensuring impactful results and measurable value for clients

Have worked for one of the largest retail chain client for delivering pricing solution for their stores.

Have been the part of team for building end to end machine learning pipeline for the client to analyze their data and provide pricing for their convenience stores across the globe.

Achieved star award from Accenture for outstanding performance from the team.

• Working for IBMs pricing product E-pricer as data scientist in pricing analytics domain .
• Contributed in building COPRA (cognitive pricing analytics) models for Epricer that are built
  on statistical prediction using LME and Quantile Regression.
• COPRA models help improve IBM and partner sales revenue and gross profit by mapping
  historical winning bid prices to current bids under consideration (Pricing Analytics).
• Developed Web service architecture that takes real time input data (XML Format) and call to
  the optimal pricing engine which uses Machine learning techniques to predict optimal price points with the help of Flask API & Postman.
• Worked with DevOps team for deployment of container-based model applications using Docker/Kubernetes on IBM Cloud.
• Built an architecture for model building pipeline and migrated model building process to workflow pipelines on cloud using Docker, Kubeflow pipelines, Kubernetes.
  Programming Language: Python
  Python packages used: Scikit-learn, Flask, matplotlib, statsmodels, faker, numpy,pandas,jaydebeapi, jpype.

• Worked for 2 years as part of Infosys AI product NIA services team after completing internship with company.
• Worked on use cases such as Text Extraction using Tesseract, building chat-bots using RASA NLU, knowledge modelling using ontology modelling as part of delivering POC’s.
  Programming Language: Python
  Python packages used: Flask, TensorFlow, pytesseract, numpy, pandas.