Hello, I am Sarvani
I'm a Java Developer
Experienced backend engineer skilled in Java development, proficient in data analysis, and adept at crafting compelling content and delivering engaging public speeches. With a passion for innovation and a commitment to excellence, I am eager to take on new challenges that require technical expertise, analytical thinking, and effective communication.
Contact meAbout Me
Experienced in analyzing the data and giving inferences for better insights.
Experienced with backend system design for seamless data flow with Kafka Pipeline.
I do content writing as a hobby
I'm enthusiastic about solving complex challenges."
I am skilled at Java
Java
Spring Boot
Kafka
Docker
ElasticSearch
Python
MS Office
Node.JS
Backend Development
Data Inference and Analysis
Public speaking and Communication
Analytial Thinking and Problem solving
Experience
Aug 2022-Present
Software Engineer
- Tarento Technologies
Aug 2022 - Present
As a backend software engineer at Tarento, I develop and integrate advanced AI solutions,
including language models and digital assistants. I lead end-to-end projects from analysis
to deployment,
optimize system performance, and handle tech migrations to platforms like Azure.
My role involves extensive work with Java, Spring Boot, Python, and various AI tools,
ensuring efficient and scalable solutions.
Spinbase : LLM replacement
Analyzed the client's existing Vector system to identify improvement opportunities and
integrate advanced language models.
Conducted end-to-end analysis from benchmarking to deployment using BART, T5, Mistral, and
Llama models, evaluating performance metrics, scalability, and alignment with client
objectives.
Provided detailed documentation, including technical specifications, competitive analysis,
and cost assessments (GCP and Hugging Face), optimizing resource allocation and maximizing
ROI.
Stack: Python, BART, T5, Llama, GCP, Hugging Face
Management Information Systems
Architected and implemented a Kafka pipeline using Java and Spring Boot for processing
data records, enabling layered data enrichment for deeper insights.
Delivered user and role-tailored visualizations, leading to a 95% increase in user
satisfaction.
Integrated Everit JSON Schema validator and JOLT Transforms, reducing invalid data entries
by 70% and improving data accuracy by 90%.
Awarded consecutive Spot Awards and a Team Impact Award for outstanding contributions.
Stack: Java, SpringBoot and Kafka, Dockers & ElasticSearch.
Tech Migration from Openstack to Azure stack
Tech Migration: Open stack to Azure stack
Worked on the migration of a Java Spring Boot application from OpenStack to Azure Stack.
Seamlessly integrated Azure Event Hubs for event streaming, replacing Kafka, Azure
Application Insights and migrated to
Azure Elastic Search.
Did End to end testing and troubleshooting the issues along with providing a comprehensive
documentation.
Stack : Azure Event Hubs, Azure application Insight, Sendgrid
THOR (AI Powered Digital Assistant ) for CX
Developed and integrated THOR, an AI-powered digital assistant for CX. Created a
fine-tuned classifier
using FAISS, trained on a large FAQ dataset for accurate customer inquiry handling.
Enhanced with OpenAI
APIs for dynamic order changes, implemented a Human-in-the-Loop framework, and designed a
flexible router
for efficient request handling, all within a week.
Stack : Python, FAISS, OpenAI API’s
AI for Predictive Maintenance
Developed and implemented an AI model for predictive maintenance,
using machine learning techniques in Python to analyze 5 years of IoT data.
Developed end to end from data preprocessing to deployment.
Employed various regression models and time series analysis techniques, performing
comparative analysis.
Integrated an alert service to notify in real-time based on model predictions,
ensuring proactive maintenance and minimizing downtime.
Stack: Python, Numpy, Scikit-learn
Education
Aug 2017- June 2022
IMTech in Information and Communication Technology
- Central University of Karnataka
220 credits
I've developed strong skills in Python, IoT, MATLAB, C, and C++, complemented by internship experiences in IoT, web design, e-commerce applications, and consultancy services. My coursework has covered subjects such as DSP, Probability and Random Process, and ACT. Additionally, I've won prizes in paper writing competitions and actively participate in workshops to enhance my knowledge
Feasibility study of drones for reliable environmental sensing IIT Roorkee
I conducted a study to develop a model of sensor-embedded drones, which act as dynamic sensors for data collection in various fields. This project presented the concept of a drone-assisted wireless sensor network, showcasing how drones can enhance data gathering and transmission. The model was developed using MATLAB, allowing for the simulation and analysis of the network's performance. Through this study, I gained insights into the integration of drones with sensor networks and their potential applications in diverse environments.
Image Processing based Photo management system IIT Palakkad
Developed an open-source system to sort images and created a pioneering deblurring model using Convolutional Neural Networks (CNN) and deep learning. The system, built using Python, achieved an impressive 88% accuracy rate in deblurring images. This project marked a significant advancement in image processing, providing a robust solution for enhancing image clarity. The work involved extensive training and fine-tuning of the CNN model to optimize its performance.
IOT based Smart Door Lock IIT Roorkee
Project was of IoT based smart door lock, to be implemented in institutions. Have worked on Nodejs, MongoDB, and Mongoose for the server communication and storing of data. Built an ExpressMail system, used SHA algorithm to protect data. Hardware was controlled using commands sent from the ESP8266-01 Wi-Fi module.
Web development intern Knockonce
Developed an E-Commerce website on OpenCart with the use of various extensions for payment gateways with GooglePay, PayTM were added. Provided a whole end to end commercial data analysis platform.
Online insustrial training on Internet of things NSIC
During my industrial internship on Industrial Applications of the Internet of Things (IoT) and applications using ESP 8266 and IFTTT, I developed a robotic car. This involved leveraging APIs and commands to enable various functionalities and automate processes. Through this project, I gained hands-on experience in integrating IoT components, programming the ESP 8266 microcontroller, and utilizing IFTTT for creating automation scenarios.
Content writing and graphic design New Delhi Connect
I have published various articles on a wide range of topics for a website blog. Additionally, I have utilized Adobe Photoshop and Adobe Illustrator to create various graphic designs and posters for the website.
Certifications
Contact Me
ysspriya14@gmail.com
Call on
+91 9999807225
Location
Bengaluru, Karnataka, India