May 2022 – Sep 2024 · 2 yrs 4 mos · Hyderabad, Telangana, India

• Data Platform:
• ● Implemented a dynamic Hot/Warm/Cold architecture in DynamoDB for handling extensive transaction data, resulting in significant cost reduction by 30%. This innovative approach optimizes resource utilization and ensures efficient scaling.
• ● Business process mining tool: Designed and developed an efficient ETL pipeline to extract, transform, and load data from various sources, ensuring data consistency and integrity. Implemented conformance and root cause analysis to identify and resolve issues within business processes, enhancing overall system reliability and performance.
• ● Designed and implemented an optimistic locking mechanism for data records to handle concurrent data access efficiently.
• ● Platform-Wide Static File Caching: Developed a comprehensive caching strategy for static files, significantly reducing AWS S3 calls. This optimization not only decreased operational costs but also accelerated response times by 50%, enhancing overall system performance and user experience.
• ● Continuous Learning ML Model Pipelines: Developed continuous learning model pipelines using Argo Workflows, automating the retraining and deployment process. This ensured models remained up-to-date with the latest data, improving accuracy and performance while reducing manual intervention.
• Video Streaming:
• ● Designed and developed a comprehensive video streaming service encompassing both on-demand and live streaming functionalities. Implemented adaptive bitrate streaming, reducing buffering times by 60% and enhancing user experience.
• ● Took complete ownership, guided engineers and actively participated in design/code reviews. Collaborated with product managers to gather requirements, provide feedback, and deliver features.

Jul 2021 – Apr 2022 · 9 mos · Hyderabad, Telangana, India

• Chat Bot:
• ● Designed, developed, tested and deployed an end to end Multi-Purpose QnA and task execution chat based bot.
• ● Automated the bot training pipeline using Argo Events, reducing manual intervention by 100%. Ensured high availability and scalability, maintaining 99.9% uptime.
• ● Deployed and managed the application on AWS end to end, leveraging services like EKS, EC2, S3. Optimized AWS resource usage, reducing operational costs by 25% while maintaining performance.

May 2020 – Jun 2020 · 1 mo

• Design of high-temperature autoclave for curing of composite laminate with Internal pressure and vacuum functionality at VASITARS Private limited
• Description - Vasitars Private Limited is a technology-based company founded by alumni of IIT Bhubaneswar. The company is governed by faculty members of Indian Institute of Technology (IIT) and Industry experts from diverse background in Pipeline technology. The project was mainly aimed to perform all the necessary analysis and calculations in the designing process of a high-temperature autoclave.
• o Analysis and evaluation for constructing suitable Industrial Autoclave for curing of composite laminate with internal pressure and vacuum functionality.
• o Thorough study which includes all necessary calculations and decisions taken for constructing the Industrial Autoclave.
• o Method of analysis includes energy calculations, parameter calculations, material selection, determination of design pressure etc.

May 2019 – Jun 2019 · 1 mo · Jodha, Orissa, India

• On-line dynamic quality and process control of rotary kilns at TATA Steel Long Products Limited to increase productivity and kiln life
• Description - Tata Steel Long Product is one of the largest specialty steel plants in India and offers a wide range of high-end steel products. The project was mainly aimed to control the quality of sponge iron produced using the data of input parameters and previous quality values of sponge iron.
• Developed a quality control algorithm based on present quality trend and input parameters using one step ahead adaptive control algorithm combined along with grey model to determine the values of input parameters in-order to maintain quality in the optimum window in MATLAB environment.
• Developed a multiple linear regression equation based on past quality and input parameters to predict the future quality trend.
• o Working closely with the higher officials of the plant to understand their requirements and developing a model.
• o Data cleansing before using it in the model.
• o Analyzing the data to choose the size of data and important parameters to be used.
• o Explaining higher officials about the progress of the work time to time.
• o Actively involved in building the code.

Dec 2018 – Dec 2018 · 0 mo · Visakhapatnam, Andhra Pradesh, India