Machine Learning Ai Engineer
Resume:
Chandra Madhumanchi
************@*****.***
Summary:
● 15 years of experience in enterprise applications development and analytical data solutions.
● Experience designing and implementing data intensive applications, end-to-end ML/AI products.
● Experience in design and build end to end ETL data engineering pipelines.
● Experience in design and code end to end ML training and inference pipelines.
● Experience with building batch and real time streaming pipelines using Spark SQL, PySpark, Kafka.
● Experience in Spark ML, Spark RAPIDS ML, fast.ai, cuML’s and MLflow.
● Experience in using Azure cloud services Event hub, stream analytics, cosmos document/graph db.’s, azure data lake store, blob storage and ADF.
● Experience in GCP services Big Query, Data Proc, Dataflow, Cloud Composer, Vertex AI.
● Experience in AWS services S3, EC2, EMR, DynamoDB.
● Hands-on Experience of Spark cluster-computing framework and Docker with Kubernetes.
● Experience with developing data intensive APIs and integrating with event driven systems.
● Experience in Flask, FastAPI, gRPC Frameworks to build microservices.
● Experience in deploying open-source models readily available within the Vertex AI in Model Garden.
● Experience in ML/DL frameworks and strong understanding of Transformer architectures, effectively develop and implement advanced Natural Language Processing (NLP) models.
● Experience in AirFlow / Kubeflow, GKE, TFX, Tensorflow serving.
● Experience with Kubernetes, YARN, Spark, and/or Ray for distributed ML orchestration.
● Excellent software engineering skills in multi-language settings leveraging Python, Java, Scala and large-scale data processing with Spark.
Certifications:
Spark - Apache Spark 2.x for Scala Certification from DataBricks Scala – Functional Programming in Scala
Python - Machine Learning Advanced Certification
Work Experience:
Company: Walmart
Title: Staff ML Engineer April 2019 – Present
● Build and deployed end to end ML pipelines for Merchandising operations with Vertex AI, PyTorch, TensorFlow, Spark ML and XGBoost.
● Designed, built and deployed scalable data ingestion pipelines using GCP services and PySpark.
● Collaborated with cross-functional teams to build and maintain a machine learning infrastructure platform, ensuring seamless integration with existing systems.
● Implemented inferencing, benchmarking, and fine-tuning of ml models, resulting in improved model accuracy and performance.
● Created ELT and data wrangling scripts to preprocess data for machine learning models, ensuring data quality and readiness for model training.
● Implemented multithreading and multiprocessing techniques to parallelize Cplex optimization process, resulting in significant improvements in model performance.
● Refactoring the notebook code to modularized and production-ready by breaking down the code into well-defined functions, organizing them into separate modules, managing dependencies clearly, and adding necessary elements like configuration management, logging, and robust error handling to ensure smooth execution in a production environment.
● Designed and built end-to-end ML pipelines to deploy and maintain optimization models on serverless Dataproc, enabling scalable and efficient model deployment.
● Developed a Spark template-based solution to load data from various sources into big query, streamlining the data ingestion process and ensuring data consistency.
● Designed and built Spark connectors to load data from Teradata, Hive into big query, enabling seamless data integration and analysis.
● Developed APIs to trigger data proc jobs for services and front-end teams, ensuring seamless integration with existing systems and enabling real-time data processing.
● Built automated business validations on model results, ensuring data accuracy and reliability.
● Deployed Hugging Face models from Vertex AI Model Garden.
● Implemented model logging and monitoring, ensuring model performance, reliability and functional monitoring.
Environment:
GCP, Vertex AI, GKE, Model Garden, Tensorflow, XGBoost, and/or PyTorch, Keras, Estimators NLTK, Gensim, fastai, Scikit-learn, SciPy, Spark MLlib, PySpark, Big Query, Cassandra, Hive, Kafka, Airflow, Azure Data Lake, NO SQL Databases, Databricks, Docker, Kubernetes, Fast API, flask and ML Flow. Company: Safeway
Title: Sr Software Engineer July 2017 – Mar 2019
Project: Recipes and Meal Planning Recommendation engine
● Build and Evaluate a recipe JSON Schema curd operation in cosmos using api’s
● Integrated third party whisk AI APIs to migrate the recipes to cosmos using api’s
● Parse ingredients data using NLP and Azure managed Language AI models api’s (LUIS’s)
● Build spark batch jobs with Scala to Load e commerce data, product catalog and purchase history to Azure SQL.
● Build Spark Structured Streaming job on DataBricks using Scala with cosmos DB change feed API to ingest recipe json’s.
● Customized Search engine on Azure Cloud
● Code and Design the cosmos octopus graph model (schema definition / optimization)
● Load the data into Graph DB using Spark Graph Frames, Thinker Pop and Gremlin queries.
● Define ARM templates for CI/CD pipelines of cosmos, Data Bricks cluster operations.
● Deployed REST API’s on PCF (Pivotal Cloud Foundry (PCF) . Environment: Azure Cloud, DataBricks Cloud, Spark Structured-Streaming, Spark Change Feed Connector, NLP, Azure Managed Language models LUÍS, Graph Frames, Spring Boot, Java, Neo4j Graph DB, Thinker Pop and Gremlin queries, Cosmos Graph DB, Document DB, SQL Server, Azure search. Company: Gap Inc
Title: Sr Data Engineer July 2015 – June 2017
Project: CDP
● Design and Created data ingestion plan for loading the data from external sources to Kafka.
● Developed custom message listener and Publisher for data ingestion pipelines for on premise Kafka.
● Created encrypt and decrypt services to perform field level encryption and decryption on fly during data ingestion for PII data.
● Set up Kafka mirroring to integrate on prem Kafka to cloud.
● Developed Spark Streaming connectors to pull data from on prem Kafka and ingest into Event Hub.
● Ingest the history data from on-prem big data 2.x cluster to event hub.
● Building the match merge layer using stream analytics.
● Building the SVOC model on cosmos DB.
● Deployed REST APIs on PCF (Pivotal Cloud Foundry (PCF) .
● Exposed API’s through Apigee and Splunk for debugging.
● Build and Deployed CI/CD pipelines for rest API’s Environment: Big Data 2.x, Azure Cloud, HDP, See Beyond, IBM MQ Kafka, Java, Spark Streaming, Event Hub, Stream Analytics, Stream Sets, Redis cache, Spring Boot and Rest API’s, Cosmos DB, Azure Blob storage, CAWA, Talend, Eclipse, Bash Script, maven, Agile, ADW, Azure data lake store and power BI. Company: HGST
Title: Software Engineer Lead April 2013 – June 2015 Project: R&D Effort Tracking and Data Model System
• Provided architecture of the system on AWS environment.
• Provided Installation & Configuration of AWS set up.
• Lead the team on technical decisions
• Understand the business requirements and translate them into technical Environment: AWS Environment, EC2, S3, SQL Server 2012, MSBI (SSIS, SSAS, SSRS) Tableau CDH Distribution, Hadoop Core, Hive, Sqoop, Kafka, Java, Oracle, MySQL Company: Cisco April 2011 – March 2013
Title: Software Engineer
Project: CEPM
● Defining architecture and implementation of distributed systems.
● Defining architecture and implementation of distributed PAP, PDP, Policy Cache Engine and replication of data using JMS.
● Wrote PL/SQL queries, functions and stored procedures. Environment: JAVA, Eclipse, Oracle 10, maven, SVN, Agile, Spring, Hibernate. Education:
Masters, Master of Computer Applications, Periyar University, India.
Contact this candidate