Yogesh Kumar

Data Scientist

New Delhi, Delhi, India8 yrs 4 mos experience

Highly StableAI Enabled

Key Highlights

Expert in building scalable machine learning solutions.
Proven track record in enhancing user engagement through recommendations.
Strong background in data engineering and cloud technologies.

Stackforce AI infers this person is a Data Science expert in E-commerce with strong capabilities in Machine Learning and Data Engineering.

Contact

Skills

Core Skills

Machine LearningData Engineering

Other Skills

Google Cloud Platform (GCP)BigQueryData PipelinesRecommendation SystemsAirflowPythonMySQLReinforcement Learning AlgorithmsArtificial Intelligence (AI)Agile MethodologiesProblem SolvingAutoMLBigQueryMLCloud StorageResponsible AI

About

Data Science Professional | NITian | Working in solving problems in E-commerce Domain| Skilled in handling data pipelines (automation), Feature Engineering on Structured and Unstructured data, Machine Learning Models, Data Crawling, Extraction, Normalization | Bachelor of Technology in Computer Science and Engineering from NIT Allahabad Recent Publications: - http://www.nhmrc2019.com.au/downloads/Parallel%20Sessions/Parallel%20Session%201C/Damien%20Bates.pdf - https://www.springerprofessional.de/en/improving-reliability-and-reducing-cost-of-task-execution-on-pre/16327862 White paper: - https://www.innoplexus.com/wp-content/uploads/2021/08/Gene-Therapy-Whitepaper.pdf

Experience

Ht digital streams

Senior Data Scientist

Feb 2025 – Present · 1 yr 1 mo · New Delhi, Delhi, India · Hybrid

Modernized HT's data & ML ecosystem by migrating all key pipelines to BigQuery and upgrading the EMR & Python environment across teams.
Built multiple recommendation engines—including similar stories, perpetual scroll ranking, and personalized feeds—driving 5–25% improvements in CTR, PVs, and D1 retention.
Introduced scalable collaborative-filtering, affinity models, and topic-based recommendation systems powering several user-facing experiences.
Implemented systemwide Airflow observability with automated alerts and self-diagnostics for data & ML pipeline failures.
Designed personalization improvements for My Feed and My HT, resulting in stronger engagement signals and better user stickiness.

Google Cloud Platform (GCP)BigQueryMachine LearningData PipelinesRecommendation SystemsAirflow+1

Purplle.com

SDE 2 - Data Science

Nov 2021 – Feb 2025 · 3 yrs 3 mos · Mumbai, Maharashtra, India

Identified the critical factors affecting the second order and first order of users at platform level using classification algorithms, feature importance and interpretability algorithms.
Deployed banner recommendation system at platform level using Thomson Sampling Bandit Algorithm and automated the same
Implemented and tested a centralised ranking recommendation system for all the merchandising properties at platform level using UCB bandit algorithm with a customised reward function to optimise Gross Margin of the platform.
Created IV (Item View) Planner at day level for top products to optimise the Gross Margin Value of the Platform for every month using Regression Algorithm and Linear Programming.
Performed A/B Testing to prove significance of results for all the experiments conducted.
Technology Used:- Python, MySQL, BigQuery, Bigtable, Airflow, Reinforcement Learning Algorithms

PythonMySQLBigQueryAirflowReinforcement Learning AlgorithmsMachine Learning+1

Innoplexus

4 roles

Associate Data Scientist

Promoted

Jan 2020 – Nov 2021 · 1 yr 10 mos

Google Cloud Platform (GCP)Artificial Intelligence (AI)Agile MethodologiesProblem Solving

Data Engineer

Mar 2019 – Dec 2019 · 9 mos

◦Feature curation and engineering from unstructured data (Clinical trials/ News/ PubMed and othersources)
◦Generating the word embeddings on clinical text data (Inclusion-Exclusion and Primary and Sec-ondary endpoints) using Gensim based Word2Vec Model and visualizing using TensorBoard
◦Knowledge of processing, cleansing, and verifying the integrity of data used for analysis
◦Working on model Interpretability and Explainability using LIME algorithm
◦Interviewing Freshers and Interns for different roles

Google Cloud Platform (GCP)Artificial Intelligence (AI)Agile MethodologiesProblem Solving

Associate Software Engineer (Financial Services)

Jun 2018 – Feb 2019 · 8 mos

◦ML based classification models on unstructured text data◦Web crawling (HTML and XML parsing)
◦Working on Information Retrieval System (Elasticsearch)◦Creating automation pipelines for web content extraction
◦Knowledge on working with external data sources/APIs (Google Custom Search Engine API and Twit-ter API)
◦Knowledge of extracting tables from HTML and classifying them into balance sheets and income state-ments based on content (German) and Calculating key figures from the tables
◦Working with SQL and NoSQL Databases and Experience in Data Modeling and Database Design
◦Working in agile based software development environment

Google Cloud Platform (GCP)Artificial Intelligence (AI)Agile MethodologiesProblem Solving

Intern Data Science

May 2017 – Jul 2017 · 2 mos · Pune Area, India

Releavance Scoring Algorithm for Information Retrieval Systems
Semantic classification models for URL classification

Artificial Intelligence (AI)Problem Solving

Motilal nehru national institute of technology

Training and Placement Representative

Mar 2017 – Nov 2017 · 8 mos · Allahabad Area, India

◦ Responsible to ensure that students are being offered high profile jobs from the campus.
◦ Communicating with the organizations from various domains of engineering and management.
◦ Monitoring and Planning the entire placement process and organization’s visit.