Resume

Sushant Kumar

Data Science & Machine Learning Enthusiast

Machine Learning & Data Science enthusiast on a mission to simplify open-source, build meaningful tech, and connect brilliant minds.

Euler Motors New Delhi, India Nov 2025 - Present

Managed and optimized 15+ Apache Flink and Spark jobs processing over 100K+ events/sec for real-time telemetry data pipelines
Designed and implemented scalable hot and cold telemetry data warehouse infrastructure using Apache Druid, Apache Iceberg, ClickHouse, and Amazon S3
Built and maintained high-throughput ingestion pipelines using Kafka Connect, AWS Glue, Athena, and Hive
Resolved small-file issues using Spark-based compaction jobs, improving query performance, storage efficiency, and long-term data reliability
Automated and optimized cron-based workflows using multithreading and batching techniques, reducing execution time by 3× and lowering AWS costs
Built monitoring and observability pipelines using Prometheus and Grafana to detect telemetry anomalies and improve production data reliability
Developed real-time alerting workflows using AWS Lambda and Amazon SNS for Kafka-to-S3 pipeline failure detection and automated incident reporting
Worked with AWS Step Functions, EMR, S3, CloudWatch, and EventBridge to orchestrate scalable stream and batch-processing workflows

Deccan AI Remote July 2025 - Present

Contributed to the evaluation and improvement of Google Gemini , enhancing accuracy, reasoning, and real-world usability
Performed large-scale annotation and systematic evaluation across domains including YouTube retrieval, messaging, utilities, and productivity
Identified performance gaps in scheduling, workflows, and user-intent understanding, delivering actionable insights for model refinement
Improved search relevance and ranking for video and song retrieval through detailed content evaluation
Detected hallucinations, factual inaccuracies, and inconsistencies to strengthen model reliability
Provided structured, evidence-backed feedback to improve response quality and consistency
Collaborated with DataCompute tools to optimize evaluation pipelines and feedback loops
Benchmarked multiple Gemini iterations to ensure consistent performance and quality improvements

Perplexity AI Remote Dec 2024 - Feb 2025

Represented Perplexity AI as a Campus Partner 2025, driving awareness and adoption within the student community
Led campus initiatives and executed outreach programs, workshops, and engagement activities
Planned and ran digital and on-ground marketing campaigns to boost visibility and user acquisition
Generated leads by onboarding students and encouraging active usage of Perplexity AI tools
Organized events ensuring smooth coordination, high participation, and strong engagement
Provided technical support and guidance to help students effectively use AI tools
Promoted Perplexity’s Comet agentic AI browser and shared actionable feedback with the team to drive product improvements
Demonstrated leadership and built strong community engagement through cross-team collaboration

Outlier AI Remote Dec 2024 - Feb 2025

Provided subject matter expertise in Biology, Chemistry, and Mathematics to improve AI model accuracy
Evaluated and refined AI-generated responses ensuring quality and scientific validity
Collaborated on improving model training pipelines and data quality
Tech Stack: Python, LaTeX, Markdown, GenAI, Mathematics, Biology, STEM

Enally.in Remote Since Feb 2021

Developed ML solutions integrating data preprocessing, model training, and deployment
Utilized Python, TensorFlow, R, Web Dev, Flask, and Google AI services
Delivered scalable AI-driven web solutions
Tech stack: Python, GenAI, Supervised Learning, R, HTML, CSS, APIs, AWS

Lovely Professional University Punjab, India CGPA: 8.05

Mount Sinai Higher Secondary School Kohima, Nagaland 72.40%

Assam Rifles Public School Kohima, Nagaland 84%

Python R SQL/NoSQL/HiveQL C C++ Scala Java Go and Rust (Basics)

TensorFlow PyTorch scikit-learn

Deep Learning

Computer Vision

Data Analysis

Apache Iceberg and Druid

GenAI & NLP

Cloud Computing

Big Data

Amazon Web Services

Oracle

NPTEL

ISRO