Adwiteey Mauriya, Ph.D.

• Utilized LangChain for LLM embeddings, enabling advanced semantic processing for text-based applications.
• Designed and built a chatbot capable of processing financial documents, allowing users to upload documents and ask relevant financial questions for immediate insights.
• Created a comprehensive feature store for the company's financial model, aimed at predicting probability of default (PD) scores, enhancing risk assessment accuracy.
• Engineered a decision model to differentiate active versus out-of-business companies, leveraging LinkedIn profile data for reliable business status classification.
• Successfully deployed, evaluated, and established the company's Business Intelligence (BI) tool on Elastic Kubernetes Service (EKS). This tool is now integral to the company's leadership and data science team for strategic decision-making and data analysis.
• Skills required: Machine learning, Natural language processing(NLP), Large Language models(LLM), Databricks, Python, Spark, Torch, Langchain

• Spearheaded the design and development of a machine learning and data pipeline for a bank account aggregator. This system integrates transaction data from over 50 global banks to calculate runway and burn rates for portfolio companies.
• Established and maintained a Continuous Integration/Continuous Deployment (CI/CD) pipeline for Extract, Transform, Load (ETL) repositories, streamlining code deployment and data management processes.
• Administered Amazon EC2 instances, ensuring optimal performance and reliability. Also actively involved in deploying and coding AWS Lambda functions for efficient serverless computing.
• Championed the implementation of code review standards, significantly enhancing code quality and collaboration. Also contributed to the backend infrastructure design, promoting scalable and robust system architecture.
• Tech stacks: Airflow, AWS, AWS Lambda, Step Functions, Python, SQL, Selenium, openAI, Large Language Models(LLM), Langchain.

• Maintained and monitored datasets in multiple data lakes and warehouses: GCP, Snowflake, Hadoop clusters.
• Supported both customer data platform and enterprise data teams.
• Optimized existing ETL processes and created new ones as required.
• Achieved cost savings of over 10 million USD annually for the team.
• Constructed datasets in BigQuery, Hive, and Snowflake.
• Championed code review standards for clean, clear, and consistent business data.
• Assisted in the onboarding of new data engineers within the team and the company.
• Tech stacks: Python, Scala, SQL, Airflow, AWS, GCP, Snowflake, Postgres, Spark, Hive

• Achieved 80% success in pilot training warehouse automation.
• Built multiple reports for the training warehouse automation.
• Produced 1.1 times the goal of qualified leads from the workshop.
• Developed flight school KPI metrics from raw company data.
• Tracked improvements in the gross margin of the company's training department.
• Tech stacks: Python, SQL, Tableau, SSRS, Excel, Power BI.

• Developed numerical tools for studying complex physical phenomena in nuclear fusion reactors.
• Utilized parallel programming in Fortran with MPI and Python.
• Scaled tools on supercomputers equipped with both CPUs and GPUs.
• Modeled fusion experiments using an underlying theoretical model.
• Developed the necessary numerical tools in Python for modeling.
• Created a Python framework to import data from fusion experiments into multiphysics codes.
• Framework facilitates various types of physics studies.
• Expertise required: Python, Scala, Mathematics, Statistics, Programming, Distributed computing, High performance computing, Physics

• Developed an algorithm and optimized it for best time complexity (efficiency).
• Implemented the optimized algorithm in a parallel HPC (High-Performance Computing) code.
• Merged the new code with an existing fix boundary equilibrium code.
• Patched the merged code into a Fortran library.
• Integrated the Fortran library with a multiphysics code.

• Provided technical support to a research group to resolve issues related to Linux operating systems.
• Assisted in setting up scientific software and libraries.
• Debugged High-Performance Computing codes.
• Consistently coded multiple batch jobs in Bash.
• Coded multiple batch jobs in Python.
• Automated various tasks through coded batch jobs.

Ph.D. in Technological Physics Engineering.
Thesis title: Effect of Impurities in Intrinsic Rotation of Tokamak Plasmas.

Master thesis in Plasma Physics.
Thesis title: Studying the tokamak equilibria using variational moment method

Major in Physics and minor in Mathematics.

Technical skills

Soft Skills

• Curious, Life long learner, Team player
• Experienced in presenting results to expert and nonexpert audiences in formal and informal settings.

Languages

English (Fluent).

Hindi (Native).

Portuguese (A2).

• UAE Golden visa holder in Talented Pioneers Category
• Top 4% in most difficult engineering exam in India (IIT-JEE) 2009.
• Received competitive KVPY fellowship in 2009(for 5 years), funded by Dept. of Science and Tech and Government of India.
• Received prestigious FCT fellowship (includes tuition fees + living expenses for 4 years) and Eurofusion research grants for participating in experiments and traveling multiple laboratories in EU.
• Obtained All India Rank 185 in India's National Ph.D. Qualifying Exam (CSIR NET).

• Analysed magneto-rotational instability using spring model analogy
• Checked energy transfer in magnetohydrodynamics turbulence using TARANG 1.0.

• Studied the Navier Stokes equation for rheological flow.
• Formulated the desired boundary conditions for simple shear flow using first order perturbation theory.