Apr 2020 – Dec 2022 · 2 yrs 8 mos · India · On-site

• ● Ingested inflight connectivity databus streams, extracting operational logs and telemetry data, and loaded structured datasets into AWS S3 and GCP Cloud Storage for analytics and ML workflows.
• ● Developed Python utilities to parse, clean, and transform connectivity and diagnostic logs, enabling structured storage in BigQuery and Redshift, reducing manual preprocessing by ~35%.
• ● Automated log extraction and dataset preparation across AWS and GCP, accelerating ML experimentation and reporting.
• ● Built Python and Java automation scripts for CI/CD build, deployment, and environment validation, cutting manual operational effort by ~30%.
• ● Supported Jenkins-based CI/CD pipelines, improving deployment repeatability and reducing release preparation time by ~25%.
• ● Assisted with containerized deployments and environment configuration across AWS and GCP, enhancing cross-environment consistency.
• ● Implemented automated monitoring and log collection using AWS CloudWatch and GCP logging, reducing troubleshooting time by ~20%.
• ● Collaborated with platform teams to validate configurations, deployment readiness, and rollback procedures for inflight software updates.
• ● Automated service validation and operational checks using Python and Robot Framework, increasing release confidence and minimizing post-deployment issues.
• ● Developed Python-based test automation integrated with Robot Framework, boosting regression coverage and reducing manual testing by ~40%.
• ● Executed functional and integration testing in simulated AWS and GCP environments, enabling earlier defect detection.
• ● Built reusable automation utilities for log validation, API verification, and environment health checks, speeding defect investigation.
• ● Participated in defect tracking, test reporting, and release validation, resolving ~80% of issues within sprint timelines.
• ● Collaborated with development teams to reproduce issues, validate fixes, and ensure CI/CD quality gates prior to deployment.

Jun 2019 – Mar 2020 · 9 mos · India · On-site

• ● Extracted and prepared telemetry data from internal databus streams, creating curated datasets for anomaly detection model training and evaluation across inflight connectivity metrics.
• ● Developed ML-based anomaly detection models using statistical and machine learning techniques, improving detection precision to ~90–93%.
• ● Applied Bayesian optimization for hyperparameter tuning, enhancing model performance, stability, and reducing manual effort.
• ● Accelerated model tuning workflows with Python multiprocessing, cutting experimentation runtime by ~40% and enabling faster ML iteration cycles.
• ● Performed feature engineering and data preprocessing on large-scale metrics, improving signal quality and reducing false positives by ~20%.
• ● Supported exploratory data analysis and validation of anomaly patterns, helping engineering teams understand connectivity behavior and recurring failure scenarios.
• ● Leveraged AWS EC2 for scalable model experimentation and batch processing workloads, facilitating efficient training and evaluation.
• ● Stored processed datasets and intermediate outputs in AWS S3 and GCP Cloud Storage, improving data accessibility and collaboration across ML and analytics teams.
• ● Utilized BigQuery for structured data warehousing and analytical queries, reducing manual data preparation time for model evaluation by ~30%.
• ● Developed reusable Python utilities for dataset extraction, transformation, and validation, improving repeatability of ML workflows.
• ● Collaborated with data engineers and platform teams to ensure reliable data availability and consistency across ML pipelines and warehouse environments.

May 2018 – Aug 2018 · 3 mos · India

• ● Developed a Python-based text extraction service to parse and process PDF documents, reducing manual document processing effort by ~40%.
• ● Built reusable parsing utilities and exposed them as a lightweight service, enhancing accessibility of extracted data for downstream analytics workflows.
• ● Implemented stock price prediction experiments using Monte Carlo simulation in R/RStudio, enabling probabilistic forecasting of market trends for analytical research.
• ● Performed data preprocessing and statistical analysis for simulation inputs, improving prediction reliability and minimizing data inconsistencies.
• ● Contributed to ML initiatives for GRC (Governance, Risk, and Compliance) risk prediction, preparing datasets and training models to identify potential organizational risk indicators.
• ● Leveraged AWS EC2 and S3 for scalable model experimentation, data storage, and batch processing, supporting efficient ML workload execution.
• ● Conducted exploratory data analysis and feature preparation for GRC datasets, improving signal quality and aiding development of predictive risk scoring models.
• ● Developed ad-hoc application components using Angular and Java to support internal tooling and data visualization, enhancing usability of analytical outputs.
• ● Assisted in debugging, testing, and enhancement of application modules, contributing to improved stability and smoother feature integration cycles.