Machine Learning Software Engineer

Juveria Fatima

Georgia,United States Email: ***************@*****.***

Professional Summary

Results-driven Software Engineer with 4+ years of experience in Python, data analytics, and machine learning model development. Skilled in solving coding challenges, writing robust test cases, and enhancing AI model capabilities. Proficient in building APIs, optimizing data access, and contributing to technical projects in remote settings.

Education

Master of Science in Computer Science (Parallel and Distributed Systems)

Osmania University, India – November 2017 GPA: 9.3/10

Bachelor of Science in Computer Science

Osmania University, India – May 2014 GPA: 8.4/10

Technical Skills

Programming Languages: R, Python, C, C++, Java, .NET

Machine Learning & NLP: Llama, LangChain, TensorFlow, scikit-learn, Transformers

Data Engineering: Databricks, PySpark, SQL, MySQL

API Development: Flask, REST APIs (Python), AWS ECS-hosted APIs

Cloud Platforms: AWS (Lambda, S3, RDS, ECS, EC2), Azure

Containerization & Orchestration: Docker, Kubernetes

Version Control & Automation: Git, GitHub Actions, Terraform

Data Visualization: Tableau, Power BI, SAS, Matplotlib, Seaborn

Content Generation & Automation: LangChain, Llama

Other Skills: Data extraction (BeautifulSoup, regex), technical writing, cross-functional collaboration

Professional Experience

Software Engineer

MSON IP – Hyderabad, India (August 2019 – July 2023)

Automated the extraction of data from patent documents using Python, leveraging tools like BeautifulSoup and regex to parse extensive datasets and store results in MySQL databases.

Designed and developed Python-based workflows using Pandas for data cleaning and processing, facilitating structured dataset access via MySQL for efficient querying.

Automated ETL processes for patent and IP data, significantly reducing manual effort by 60% and increasing the efficiency of data integration. This automation provided faster and more reliable access to critical data for patent analysis, supporting quicker decision-making across legal and R&D teams.

Developed complex SQL queries in Azure Databricks and Snowflake, optimizing data retrieval by reducing query time by 25%, which enhanced patent data accessibility for machine learning models and improved classification accuracy. These optimizations allowed stakeholders to access insights more efficiently, directly supporting patent classification and strategic IP decisions.

Engineered a data layer automation tool using Python and SQL, reducing data processing time by 45% and improving dashboard load times. This enriched the end-user experience by providing instant access to patent insights, which accelerated patent analysis and enhanced IP portfolio management.

Built interactive dashboards in Tableau and Power BI, delivering actionable insights on patent trends, classifications, and IP performance. These visualizations improved decision-making speed by 35%, enabling teams to quickly assess patent data and align IP strategies with organizational goals.

Applied predictive analytics using NumPy, Pandas, SciPy, and Scikit-Learn, increasing demand forecasting accuracy by 15% and improving model performance by 20%. These insights allowed for more accurate inventory and resource planning, ensuring better alignment with patent filing trends and anticipated R&D needs.

Developed and deployed Python APIs using Flask for real-time patent classification, enabling legal and R&D teams to classify patents more quickly and accurately. This real-time functionality improved operational efficiency, ensuring end-users could rely on up-to-date patent data.

Designed and implemented cloud-based data pipelines using AWS S3 and Lambda for secure data storage and on-demand data transformation, automating data feeds into machine learning models. This cloud solution enabled scalable and efficient processing of patent data, ensuring continuous availability of up-to-date insights to support IP evaluations.

Coop Engineer

Advanced Micro Devices – Hyderabad, India (November 2016 – July 2017)

Conducted research on Convolutional Neural Network (CNN) models such as AlexNet, VGGNet, and LeNet, gaining insights into their architectures and applications for image detection in autonomous driving systems.

Developed Python-based scripts using TensorFlow and Keras to train CNN models, including VGGNet, on custom datasets, improving model performance for object detection tasks.

Created data generation pipelines with TensorFlow and Keras to augment training datasets, enhancing the model's ability to generalize across various scenarios encountered in autonomous driving.

Developed C and C++ scripts to implement foundational convolutional operations, such as convolution, pooling, and ReLU, facilitating the simulation and testing of CNN layers for efficient computation.

Contributed to the project "Tiny Hardware Accelerator for Image Detection Using Deep Learning," focusing on designing a lightweight hardware solution to accelerate image detection tasks, enhancing performance for real-time applications.

Collaborated closely with the R&D and VLSI teams to integrate CNN logic into edge devices, supporting the development of an embedded solution for real-time decision-making in autonomous vehicles.

Assisted in deploying CNN-based deep learning models on hardware platforms, enabling optimized model inference and real-time responsiveness in self-driving applications.

Software Developer

Andhra Pradesh Police Headquarters – Hyderabad, India (May 2011 – November 2011)

Led a team to design and develop a VB.NET application interfacing with SQL databases to automate the retrieval and display of police officials' rank and medal information.

Managed project timelines and priorities, ensuring on-time delivery of a user-friendly application that streamlined internal data management processes.

Projects

AI-Based Blog Generation using LangChain and Llama

Developed a blog generation system using LangChain and the Llama language model to automate the creation of SEO-optimized content.

Managed the full model lifecycle, from data ingestion to model deployment and monitoring, ensuring reliable performance over time.

Leveraged Llama's advanced natural language understanding capabilities to generate coherent and topic-specific blogs, integrating LangChain for workflow automation.

Deployed the solution on AWS Lambda and S3 for scalable, serverless execution, minimizing infrastructure costs and ensuring seamless operation.

Patentability Analysis: Developed APIs for preprocessing and analyzing large patent databases using Python and SQL, extracting features for machine learning models. Designed cloud-based pipelines on AWS for automated data processing and model integration.

Criminal Face Detection System: Created a Java-based system with facial recognition algorithms to support real-time law enforcement applications.

Scalable and Secure Sharing of PHR: Built an Android application using Attribute-Based Encryption to ensure secure and scalable sharing of personal health records.

Certifications

AWS Cloud Practitioner

Contact this candidate