Experience

A comprehensive history of my professional engineering and research roles.


I am authorized to work in the United States for any employer under OPT / STEM OPT [3 years remaining].

Experience

AI Automation Engineer

Marketeq DigitalMiamiRemoteMiami Noir LandscapeMIAMiamiLOC // MIAFlorida, USACOORDS25.7617° N, 80.1918° W

Built a multi-tenant AI automation service for Marketeq's outreach offering, integrated into their lead-intelligence platform — taking social engagement from content generation to lead conversion.

  • •Re-architected an early prototype into a production microservices platform supporting multiple users and social channels.
  • •Designed and deployed on Google Cloud (Cloud Run + S3) as a multi-tenant system with a platform-agnostic scheduler, so new channels onboard without touching the core pipeline (LinkedIn live).
  • •Built a multi-agent, multimodal content-generation system (Gemini + Anthropic APIs) for text and images, with Playwright-based LinkedIn automation.
Google Cloud RunAWS S3PlaywrightGemini APIAnthropic APIMicroservicesMulti-Agent

AI / ML Engineer Intern

LivePerson Inc.New YorkNew York, NYNew York Noir LandscapeNYCNew YorkLOC // NYCNew York, USACOORDS40.7128° N, 74.0060° W

Built prompt optimization systems to enable quick and efficient migration to different LLMs and LLM providers.

  • •Architected automated prompt evaluation and optimization frameworks across leading LLMs.
  • •Streamlined model evaluation workflows, reducing migration friction and token latency.
PythonLLMsPrompt OptimizationDSPyFastAPI

Machine Learning Research Intern

Center for Information Security, Forensics and Cyber Security Research (C-ISFCR) @ PES UniversityBengaluruBengaluru, IndiaBengaluru Noir LandscapeBLRBengaluruLOC // BLRKarnataka, IndiaCOORDS12.9716° N, 77.5946° E

Real-time video analysis for anomaly detection using deep learning models.

  • •Researched spatio-temporal deep learning pipelines for detecting behavioral anomalies in high-throughput video streams.
  • •Optimized model inference latency for edge deployment and surveillance data feeds.
PyTorchComputer VisionDeep LearningVideo Analysis

Founding Software Engineer

Param Innovation (The Param Science Experience)BengaluruBengaluru, IndiaBengaluru Noir LandscapeBLRBengaluruLOC // BLRKarnataka, IndiaCOORDS12.9716° N, 77.5946° E

Developed interactive web and AI applications for the Param Science Museum, enabling visitors to explore scientific concepts through interactive AI and ML models.

  • •Designed and shipped interactive museum installations blending physical exhibits with real-time computer vision.
  • •Engineered scalable web applications and interactive demo interfaces for thousands of visitors.
PythonComputer VisionReactNode.jsInteractive Media