Harsh Maheshwari Email and Phone Number
Harsh Maheshwari work email
- Valid
Harsh Maheshwari personal email
I am an IIT Bombay graduate Data Scientist with undergraduate degree in chemical Engineering and 2 years of hands-on experience designing, building and managing the complex production lifecycle of end-to-end machine learning pipelines. I am a self managed individual who can manage multiple stakeholders and get the required context to conceptualise goals and implement solutions tailored to the market situation. I hone the ability to scope out a task and implement a balanced solution taking into consideration the computational complexity and other restrictions. Always trying to stay current on the latest machine learning research relevant to the problems at hand. I take pride in leading by showing and knowing how. Ready to roll up the sleeves and do anything to get the job done and well - no task is menial or tedious. Do visit my portfolio at harsh-maheshwari.github.ioHands on Experience in Deep Learning and Machine Learning- Supervised Learning: Linear and Logistic Regression, Gradient Boosting Machines (XGBoost, LightGBM, CATBoost), Random Forests, Support Vector Machines- Unsupervised Learning: K-means Clustering, Generative Adversarial Networks.- Additional Architectures: CNNs, U-Net, RNNs (LSTM, GRU), Transformers, BERT, Temporal Attention, Multi-Head Attention, Hierarchical Attention- Embeddings and methods : Word2Vec, Doc2Vec, Glove- Dimensionality Reduction : TSNE, Principal Component Analysis, Clustering- Probability Distribution Models : Gaussian Mixture Models, Gamma Models etc - Nonparametric Methods I have strong foundations in the mathematical aspects including Linear Algebra, Probability Theory, Statistical Modelling, Analysis of Variance, Convex Optimisation, Hypothesis TestingBasic Theoretical Understanding in Reinforcement Learning: Q-learning, Deep Q-Networks, Deep Deterministic Policy GradientTechnology stack:Tensorflow 2.X Universe, Tensorflow Extended (TFX), Kubeflow, Apache Beam, Dataflow & Kubernetes, Unix Shell Scripting, Python, Sklearn, OpenCV, Pytorch, Torch, Numpy, Pandas, PySpark, Matplotlib, Seaborn, Ggplot, Tableau, SQL, PostgreSQL, NOSQL, MySQL, Apache Spark, Git, AWS, GCPOpen Source Contributions: Kedro
Nk Securities Research
View- Website:
- nksecurities.com
- Employees:
- 125
-
Quantitative ResearchNk Securities ResearchGondia, Mh, In -
Data Science ManagerGoglocal Mar 2023 - Oct 2023Developed multi-channel e-commerce strategies for cross-border trade. Scaled the system to 1000+ SKUs utilizing advanced data scraping and machine learning to ensure a tremendous reduction in human effort.Leading the cloud deployment and scaling using Ray of a merchant-facing product that generates actionable plans for clients using machine learning and deep learning, including sales forecasts, inventory management, and expected profit margins on a per SKU level, across major global… Show more Developed multi-channel e-commerce strategies for cross-border trade. Scaled the system to 1000+ SKUs utilizing advanced data scraping and machine learning to ensure a tremendous reduction in human effort.Leading the cloud deployment and scaling using Ray of a merchant-facing product that generates actionable plans for clients using machine learning and deep learning, including sales forecasts, inventory management, and expected profit margins on a per SKU level, across major global markets such as Amazon, eBay, Walmart, and Lazada.Empowered clients to improve their marketplace performance by developing performance metrics for merchant SKUs, comparing them against similar SKUs using natural language techniques like attribute extraction and semantic similarity. Scaled the algorithms to work on various marketplaces, resulting in optimized listings and improved sales. Built product listing enhancement tools using keyword extraction and Natural language generation and Background generation using Stable Diffusion modelsSpearheaded the development & and scaling of scraping pipelines for Amazon websites in multiple countries, extracting information to drive insights & and inform decision-making. Show less -
Data ScientistBlue Yonder Apr 2022 - Mar 2023Part of a 15+ member team working on enabling intelligent decisions via cloud native SaaS ML solutions in all areas of supply chain for omnichannel commerce like Store fulfillment capacity, Delivery date estimation (pick, pack, ship and transit times), Demand Forecasting, Inventory Allocation, Dynamic safety stock, Order fulfillment probability, Markdown, Stockout avoidance etc.- Working on massive deep learning global models on TensorFlow and Keras around TensorFlow Extended pipelines… Show more Part of a 15+ member team working on enabling intelligent decisions via cloud native SaaS ML solutions in all areas of supply chain for omnichannel commerce like Store fulfillment capacity, Delivery date estimation (pick, pack, ship and transit times), Demand Forecasting, Inventory Allocation, Dynamic safety stock, Order fulfillment probability, Markdown, Stockout avoidance etc.- Working on massive deep learning global models on TensorFlow and Keras around TensorFlow Extended pipelines on Kubeflow with data streaming & processing in Apache Beam on Dataflow.- Deep experience in time series analysis, statistics, network analysis, and machine learning / deep learning with large and noisy datasets (> 5TB)- Using an auto scaled kubernetes platform and with a multi-tenancy solution to share cloud resources for better utilization and improved margins- Ability to condense and articulate outcomes of analyses for non-specialized audiences.- Self-motivated team player with extensive international experience and proven take charge ability. Show less -
Technology Graduate In Enterprise Data And Intelligent AutomationAnz Jun 2021 - Mar 2022Bangalore Urban, Karnataka, India -
Business Intelligence InternAxis Bank Apr 2020 - Jun 2020IndiaBuilt a micro-service for predicting customer income using nominatim API, scikit-learn, docker and kubernetesEngineered new featurebased on customer address using K nearest neighbor algorithm and geocoding services Obtained accuracy of over 90% using a XGboost regression model based on customer history and transaction data Developed a web interface using Django framework and containerized in Docker container for easier deployment Reduced daily up time by using… Show more Built a micro-service for predicting customer income using nominatim API, scikit-learn, docker and kubernetesEngineered new featurebased on customer address using K nearest neighbor algorithm and geocoding services Obtained accuracy of over 90% using a XGboost regression model based on customer history and transaction data Developed a web interface using Django framework and containerized in Docker container for easier deployment Reduced daily up time by using Kubernetes orchestration for automated managing and scheduling of services Show less -
Computer Vision LeadInnovation Cell, Iit Bombay Jul 2019 - Jun 2020Mumbai Area, IndiaDesigned and developed an autonomous fixed wing for search and rescue missions customized for Barselona Smart Drone ChallengeIntegrated a ROS enabled camera system for detection of characters from a height of 50 meters on a flying droneDeveloped a multi-core zoom in character recognition model based on Char74 dataset using OpenCV & KerasObtained accuracy of 94% using transfer learning on MobileNetV2 architecture on custom aerial data set -
Research In Artificial Intelligence InternNus Computing Nov 2019 - Jan 2020SingaporeDeveloped a Temporal Attention modelfor node classification and link prediction tasks in different temporal graphFound efficiencyof models like Attention Walk, TMF, CTDNE, BANEon different datasets for link prediction taskUnderstood dependency of graphs on time using dynamic models like Node2VecandTemporal Node EmbeddingFound inefficienciesin the above models and improved them to get AUC of 86%for College Messages datase -
Data Analyst InternPv Diagnostics Nov 2018 - Dec 2018Mumbai Area, IndiaDesigned SQLite data base, derived new insights and detected outliers through Data Sorting, Cleaning and EDAClassified different type of defects in solar panels using Logistic Regression Model, achieved 85% accuracyDeveloped new parameters of physical significance to a solar plant based on the translation equation
Harsh Maheshwari Skills
Harsh Maheshwari Education Details
-
Chemical Engineering -
Chemical Engineering
Frequently Asked Questions about Harsh Maheshwari
What company does Harsh Maheshwari work for?
Harsh Maheshwari works for Nk Securities Research
What is Harsh Maheshwari's role at the current company?
Harsh Maheshwari's current role is Quantitative Research.
What is Harsh Maheshwari's email address?
Harsh Maheshwari's email address is ha****@****ies.com
What schools did Harsh Maheshwari attend?
Harsh Maheshwari attended Indian Institute Of Technology, Bombay, Dtu - Technical University Of Denmark.
What skills is Harsh Maheshwari known for?
Harsh Maheshwari has skills like C++, Data Analysis, Business Analysis, Keras, Matlab, Autocad, Deep Learning, Machine Learning, Front End Development, Data Mining, Statistics, Pytorch.
Who are Harsh Maheshwari's colleagues?
Harsh Maheshwari's colleagues are Mahesh Nimbal, Mahendrasingh Chundavat, Deepak Gurjar, Faraz Khan, Anup Kumar Sharma, Yatin Dutt Gupta, Sheldon Mathews.
Not the Harsh Maheshwari you were looking for?
-
Harsh Maheshwari
Digital Alliance @ Rgi | Strategy & Scale | Insurance | Fintech | Ba | Pm | Ex-Flipkart | Ex-Flexiloans | Ex-ShadowfaxBengaluru -
Harsh Maheshwari
@Thebrivo | Crafting Digital Success Through Innovative Marketing Strategies! 🚀 Let’S Connect And Elevate Your Brand Together. Dm For Collaboration! 📈Beawar -
Harsh Maheshwari
3X Aws Certified | Ai & Cloud Enthusiast | Architecting Scalable Solutions For Cloud TransformationGurugram -
Harsh Maheshwari
Software Engineer At V2Force | V2Solutions | 1XsalesforcecertifiedRajasthan, India -
Harsh Maheshwari
Pune
Free Chrome Extension
Find emails, phones & company data instantly
Download 750 million emails and 100 million phone numbers
Access emails and phone numbers of over 750 million business users. Instantly download verified profiles using 20+ filters, including location, job title, company, function, and industry.
Start your free trial