Nitesh Chandnani

AI Researcher

Bengaluru, Karnataka, India8 yrs 4 mos experience

Most Likely To SwitchHighly Stable

Key Highlights

8 years of experience in AI & Machine Learning.
Proven track record in developing scalable ML solutions.
Expert in bridging technical implementation with practical use cases.

Stackforce AI infers this person is a Data Science professional with expertise in Machine Learning and AI across various industries.

Contact

Skills

Core Skills

Machine LearningData ScienceData Analysis

Other Skills

A/B TestingAnalyticsArtificial Intelligence (AI)AzureBERTClusteringComputer VisionData AnalyticsData ProcessingData VisualizationDebuggingDeep LearningHyperparameter TuningImage ProcessingMachine Learning Techniques

About

Data science professional with 8 years of hands-on experience with AI & Machine Learning to create & deploy efficient & accurate models. Experienced at creating statistical & deep learning algorithms using predictive data modeling, mining & processing to deliver data-driven solutions. Skillful at working with product teams to bridge the gap between products' practical use cases & their technical implementation using mathematical algorithms. Apart from work, I like to read fictional novels and write poetry. It gives me immense pleasure to convert my thoughts into words chosen from English, Hindi or Urdu languages and put them in front of the people on my personal WordPress blog page. I also enjoy indulging in singing and have an apt interest in multiple sports like cricket, badminton & lawn tennis.

Experience

Visa

Senior Data Scientist

Apr 2024 – Present · 1 yr 11 mos · Bengaluru, Karnataka, India · On-site

Working towards developing scalable ML solutions for North America clients

Inmobi

Senior Data Scientist

Jul 2023 – Mar 2024 · 8 mos · Bengaluru, Karnataka, India · On-site

Led the pricing charter to enhance ML-driven bidding for ~500M daily adrequests, covering model development, A/B testing, & platform rollout
Improved LightGBM CTR & CVR models by adding user-level counters & newfeatures to get 14% offline AUPR lift & 1.5% install value lift on iOS & Android
Performed hyperparameter tuning for the same model via Hyperopt in PySpark,boosting offline AUPR by 10% & install value by 0.7% post full rollout

Machine LearningA/B TestingModel DevelopmentHyperparameter TuningPySparkData Science

Algonomy

2 roles

Senior Data Scientist

Promoted

Jan 2021 – Jul 2023 · 2 yrs 6 mos

Developed & deployed a multiclass DETIC model to identify product category & color from images, achieving micro recall scores of 0.6 (category) & 0.4 (color)
Built a Random Forest model using CLIP image embeddings to detect & identify entities in apparel images, achieving a micro F1 score of 0.84
Built a BERT-based topic model for product tagging, optimized keywords to remove similar tags using PolyFuzz, & achieved a coherence score of 0.25

Deep LearningComputer VisionRandom ForestBERTImage ProcessingMachine Learning+1

Data Scientist

May 2019 – Dec 2020 · 1 yr 7 mos

Built a Markdown Optimization model to predict optimal discounts & maximize profit, using L1-L2 regularization & gradient descent to achieve MAPE of 30%. Implemented parallelization to utilize multiple cores, reducing model training time by 75%.
Developed a Customer Lookalike model using a binary XGBClassifier, achieving an F1 score of 0.7 using Bayesian Optimization
Developed functions to read data directly from Azure/Amazon S3, enabling 3x faster data fetching and processing by decoupling from SQL queries

Machine LearningXGBoostData ProcessingAzureSQLData Science

Ola (ani technologies pvt. ltd)

Analyst

Dec 2017 – Apr 2019 · 1 yr 4 mos · Bengaluru Area, India

Created a Random Forest model for personalized cab category recommendationswith 95% accuracy, displaying the top 3 preferences
Implemented a DBSCAN clustering algorithm to identify pickup location clustersreducing cancellations by 3% & increasing daily platform earnings by upto ₹ 3.5L
Used pandas, numpy, and SQL to cache frequent user search strings for popular places in 9 major cities (India & Australia), saving Rs. 40,000 per day

Random ForestClusteringSQLData AnalysisMachine Learning

Fuzzy logix

Associate Data Scientist

Jul 2017 – Nov 2017 · 4 mos · Bengaluru, Karnataka, India

Worked mainly on exploring and understanding the algorithmic aspect of mathematical functions used in DBLytix, Largest selling product of Fuzzy Logix, & debugging them to make them really fast & memory efficient
Explored the mathematical view behind major machine learning techniques like regression, decision trees, Random Forests, SVMs, ensemble learning, boosting, deep learning etc. & studied the areas of their application

Mathematical FunctionsDebuggingMachine Learning Techniques