BExperience
Work that shipped.
AI agents on Snowflake, IoT data pipelines and autonomous underwater robotics. Each role with its actual scope, stack and verified outcomes.
Jan 2026 - Jul 2026
USA
- AI agents migrated to production
- 18
- AI agents migrated to production
- Snowflake views audited
- 200+
- Snowflake views audited
- agents integrated with Salesforce CRM
- 5
- agents integrated with Salesforce CRM
- Developed Snowflake Cortex AI agents, SQL stored procedures and Model Context Protocol (MCP) integrations to automate financial reporting, so teams retrieve accurate revenue insights without manual data analysis.
- Migrated 18 AI agents to production and audited 200+ Snowflake views.
- Implemented OAuth, role-based access control (RBAC) and REST APIs to protect customer data and keep reporting secure and reliable.
- Integrated Salesforce CRM APIs with five AI agents.
- Built automated SQL and JSON validation tests to verify data accuracy, troubleshoot integration failures and keep financial reporting reliable.
- Built frontend features and REST API integrations for financial reporting, working with senior engineers to review changes and resolve integration issues.
- Snowflake Cortex
- SQL
- Stored procedures
- MCP
- OAuth
- RBAC
- Salesforce APIs
- REST APIs
Jan 2024 - Aug 2024
Chennai, India


Data Analyst & Engineer
Flare Innovation
Pipelines and dashboards for IoT operations data, with quality checks that stopped bad records before reporting.
- IoT records through Python ETL
- 1M+
- IoT records through Python ETL
- Developed Power BI dashboards integrating PostgreSQL and Airtable APIs to track operational KPIs and improve reporting efficiency for engineering and business teams.
- Engineered Python ETL pipelines processing 1M+ IoT records using PostgreSQL and Airtable, with automated PyTest schema validation to improve data quality.
- Improved device reliability through Python-based sensor anomaly detection and SQL transformations, using PySpark to optimize large-scale data processing.
- Implemented GitHub Actions CI/CD pipelines for automated testing and builds, containerizing data workflows with Docker for consistent, reliable deployments.
- Python
- SQL
- PostgreSQL
- Airtable API
- Power BI
- PySpark
- PyTest
- Docker
- GitHub Actions
Jun 2022 - Oct 2022
Chennai, India


Machine Learning Engineer
Sri Sairam Techno Incubator Foundation
Perception and control for an autonomous underwater vehicle, from simulation to the pool.
- of 320+ teams, Singapore AUV Challenge
- 6th
- of 320+ teams, Singapore AUV Challenge
- success rate in sim and pool tests
- 92%
- success rate in sim and pool tests
- Developed autonomous underwater navigation and object detection for the team's AUV, contributing to a 6th-place global finish among 320+ teams at the Singapore AUV Challenge.
- Trained and integrated YOLOv5 models with OpenCV for real-time underwater object detection.
- Developed reinforcement-learning control integrated with the Robot Operating System (ROS), reaching a 92% success rate on obstacle avoidance, gate traversal and object recovery across Gazebo simulations and real pool tests.
- Python
- YOLOv5
- OpenCV
- Reinforcement learning
- ROS
- Gazebo