Shenoy Pratik Gurudatt

About

I build observability and search tools used by thousands of developers worldwide.

Currently an SDE at AWS, maintaining OpenSearch - one of the most active open-source search and analytics engines. My work spans the OpenTelemetry ecosystem, data ingestion pipelines, and developer-facing dashboards.

Previously: DeepFake detection research at USC-ISI with published work at ECCV, and computer vision for retail at SAP Labs with 2 granted US patents.

Outside of code - I love traveling, discovering new cuisines, and exploring music from hip hop to dubstep.

opensearch · opentelemetry · observability · search · machine learning · opensearch · opentelemetry · observability · search · machine learning · opensearch · opentelemetry · observability · search · machine learning · opensearch · opentelemetry · observability · search · machine learning · opensearch · opentelemetry · observability · search · machine learning · opensearch · opentelemetry · observability · search · machine learning ·

Open Source

0

PRs merged to OpenSearch

0

Contributions to OpenTelemetry

0

GitHub Orgs (AWS, OpenSearch, OTel)

Contribution Activity

Featured Repositories

Observability·APM·Data Prepper·OTel Collector·Search Relevance·Navigation·CVE Fixes·Helm Charts· Observability·APM·Data Prepper·OTel Collector·Search Relevance·Navigation·CVE Fixes·Helm Charts

View all contributions

Experience

2021 - PRESENT

Software Engineer

Amazon Web Services · OpenSearch

Maintainer & contributor to the OpenSearch Project. Building observability features, search relevance tools, and APM dashboards. Leading icon-based navigation overhaul across plugins. Active community engagement with conference talks and open-source advocacy.

TypeScript Java React Kotlin OpenTelemetry

2019 - 2021

Graduate Research Assistant

USC · Information Sciences Institute

Developed deep learning models for DeepFake detection under Dr. Wael AbdAlmageed. Published at ECCV 2020. Built pipelines to distinguish GAN-generated images from real ones using novel two-branch recurrent architectures with Laplacian of Gaussian bottleneck.

Python TensorFlow PyTorch OpenCV

2016 - 2019

Machine Learning Engineer

SAP Labs · Bangalore

Designed computer vision pipelines for product detection and recognition in retail. Generated thousands of training images via augmentation. Built continuous evaluation pipelines and published ML APIs to production at scale. Co-invented 2 US patents in neural fabric architectures.

Python OpenCV Flask Docker

Speaking

MEMBER OpenSearch Observability TAG

Technical Advisory Group member shaping the future of open-source observability. Advising on metrics, logs, traces, and profiles storage - collaborating with OpenTelemetry, Prometheus, Jaeger, and Fluent Bit. Members include engineers from AWS, Apple, SAP, and Uber.

OpenSearchCon India 2026 Mumbai, India

Agent Under the Microscope: Monitoring Agentic Workflows with OpenSearch

Introducing Agent Traces and Agent Health for OpenSearch - a native UI for exploring agent execution traces using OTel SDKs with GenAI semantic conventions, hierarchical trace views, and golden path comparison.

OpenSearchCon India 2026 Mumbai, India

From Queries to Insights: AI-Powered Database Monitoring with OpenSearch and Prometheus

Monitoring production databases on Kubernetes using OpenSearch and Prometheus with an OTel-based pipeline, AI-assisted investigation via text-to-PPL and text-to-PromQL for natural language querying.

SCaLE 23x · 2026 Pasadena, California

OpenSearch: The Open Source Path to Search and Observability

Introduction to OpenSearch - from indexing unstructured logs to full observability with tracing, monitoring, vector search, and NLP capabilities. Covering the project's journey to The Linux Foundation.

OpenSearchCon NA 2025 San Jose, California

From Kernel to Cloud: Building a Complete Open Source Observability Pipeline

Full-stack observability using eBPF instrumentation via Inspektor Gadget, unified through OpenTelemetry Collector, analyzed with OpenSearch and Prometheus (Cortex). Featuring PromQL integration.

Haystack 2022 Search Relevance Conference

OpenSearch - Ecommerce Search & Discovery Platform Powered by Querqy

Building a personalization platform for e-commerce search using Querqy - optimizing search relevance with boosting, typo tolerance, synonyms, and dynamic filtering via OpenSearch Dashboards.

View talk details

View all sessions on Sessionize

Writing

Apr 2026 opensearch.org

Data Prepper 2.15 brings Apache Iceberg ingestion and Prometheus metrics support

Announcing new capabilities in Data Prepper including Apache Iceberg sink for data lake workflows and native Prometheus metrics ingestion.

Sep 2025 opensearch.org

Announcing OpenSearch Observability TAG: Shaping Open Source Observability Together

Introducing the Technical Advisory Group for Observability, bringing engineers from AWS, Apple, SAP, and Uber to shape the future of open-source observability.

Aug 2025 opensearch.org

Diving deep into distributed microservices with OpenSearch and OpenTelemetry

End-to-end guide on tracing distributed microservice architectures using OpenTelemetry instrumentation with OpenSearch as the backend.

Jul 2025 opensearch.org

Announcing OpenSearch Data Prepper 2.12: Additional sources and sinks

Release announcement for Data Prepper 2.12 with new data ingestion sources and sink destinations for flexible pipeline configurations.

View all posts on opensearch.org

Education

M.S. Computer Science

University of Southern California

Specialization in Computer Vision, Deep Learning, Databases, and Data Mining. Research projects in AI algorithms, Big Data, and Deep RL for game automation.

Research Assistant at USC-ISI · Best Data Science Developer Award (CKIDS 2019)

B.Tech. Computer Science

SRM University, Chennai

Core subjects: Software Engineering, Soft Computing, Object Oriented Programming, Human Computer Interaction. Bachelor thesis on wireless communication using DTMF tones.

Microsoft Imagine Cup National Finalist · 1st Place Worldwide UX Challenge

Certifications

CNNs in TensorFlow - Coursera Intro to TensorFlow for AI/ML/DL - Coursera NLP in TensorFlow - Coursera

Research &
Patents

ECCV 2020 European Conference on Computer Vision

Two-branch Recurrent Network for Isolating Deepfakes in Videos

Iacopo Masi, Aditya Killekar, Royston Marian Mascarenhas, Shenoy Pratik Gurudatt, Wael AbdAlmageed

A two-branch network architecture that amplifies artifacts while suppressing face content using a Laplacian of Gaussian bottleneck layer. Evaluated on FaceForensics++, Celeb-DF, and Facebook DFDC benchmarks.

US10474928B2 · Granted

Dynamic Self-Learning System

A neural fabric architecture using Monte Carlo tree search to dynamically create lateral flows between learned CNN patterns for product classification.

SAP SE · Nov 2019

US10467501B2 · Granted

Computer Vision Architecture with ML Image Recognition

Multi-hierarchy classification system using smart contour models (deep CNNs) to identify product shapes and route through shape-specific classifiers.

SAP SE · Nov 2019

Recognition

🎖️

Best Data Science Developer

USC-CKIDS × GRIDS

2019

⭐

Innovation Award

SAP Labs

2018

🥇

1st Place Worldwide

Microsoft Imagine Cup - UX Challenge: World Citizenship

$3,000 Prize · 2015

🏆

National Finalists

Microsoft Imagine Cup - Top 29 teams in India

2015

Stack

Languages

TypeScript JavaScript Python Java Kotlin Go C

ML / AI

TensorFlow PyTorch OpenCV scikit-learn MLFlow

Cloud / Infra

AWS OpenSearch Kubernetes Docker Kafka OpenTelemetry

Frontend

React HTML/CSS Tailwind Angular

Tools

Git GitHub Actions Grafana Helm Data Prepper

Connect

Let's build something together.