Hello, It's me

Shiva Sai Charan

I'm a

A passionate Engineer curious about how systems work and having an experience building scalable services leveraging various tech and cloud. I learn through by doing cool projects.

Shiva Sai Charan's Avatar Image

skills

experience

Research Assistant

Northeastern University

  • Assisted students at AiSunks club get their hands on the open source projects. Organized meetings and monitor work progress.

  • I have been editor for AiSunks medium magazine and github pages.

January 2023 - May 2023

Northeastern University

Software Engineer (Golang)

ZopSmart

  • Developed customer personalization API in Golang resulting in improved targeted marketing. These services scaled to handle a throughput of 9k requests/min.

  • Optimized database queries resulting in user details API response time reduced by 5% leading to a 20% decrease in resource utilization.

  • Migrated application deployments from Spinnaker to Harness with a quick adoption to a new business deal.

  • Revamped the UI by adding new functionalities for an internal tool, have improved usability and bring down testing efforts by 30%, effectively resolving data inconsistencies across various services.

  • Enhanced locking mechanism to handle bulk records helped migrating 100M records faster by 20%.

  • Performed 30+ PR reviews, Handled 6+ Production releases and given a couple of knowledge transfers making new folks on par with the rest of the team.

May 2021 - June 2022

ZopSmart

Software Engineering Intern

ZopSmart

  • Developed API's in Golang by adapting to layered architecture and documented requirements in swagger.

  • Integrated Redis-based Session management to online examination service. This lead to a 5% reduction in Session disruptions, thereby improving exam reliability and enhanced overall user experience.

  • Orchestrated CI/CD pipelines to streamline the process of unit testing, code containerising, publishing build images to Registry and deploying to GCP has reduced delivery time by 80% and ensuring consistent releases.

June 2021 - May 2021

ZopSmart

SWE Intern

Jivass Technologies

  • Scrapped and curated 3000 train schedule data.

  • Developed end to end application to visualize trains network on the graph using D3.js and simulate trains according to their schedule.

  • This project is a proof of concept for algorithms that are designed/analyzed to minimize train delays.

  • Deployed the application on E2E network.

May 2020 - Jul 2020

Jivass Technologies

projects

Assignments Tracker

Assignments Tracker

This Project is an attempt to learn to setup a easily reproducible and consistent infra on AWS for an small scale application like assignments submission portal considering every aspect of security, availability, reliability, scalability leveraging various cloud services.

#Golang

#AWS

#IAM

#EC2

#S3

#VPC

#Subnets

#Internet Gateway

#Route Table

#Route53

#Lambdas

#Security Groups

#RDS

#Parameter Group

#Cloud Watch

#StasD

#Alerting

#Load Balancers

#Auto Scaler

#DNS

#Name Servers

#CAA Records

#SubDomains

#Key Vault

#Git Actions

#Packer

#SystemD

#Shell Script

#SSl Certificates

#Roles & Policies

#Amazon Simple Mailing Service

Audio Brief

Audio Brief

Designed data pipelines to process 5k ebooks from the dataset to extract each chapter and summarize chapters, which were then converted to audio books that are stored to google cloud buckets. Engineered the process of chapter extraction through Google Pub/Sub and Cloud Functions, resulting in optimized efficiency and enhanced reliability. Leveraged LLM with langchain and Google TTS as a Model as a Service to deliver on demand summaries and audio transcripts for selected chapters from a book. Enhanced application debugging by facilitating proper log messages and metrics aggregated using Cloud Logging.

#Python

#FASTAPI

#Multi-Threading

#Airflow

#Pinecone

#GCP

#Cloud Storage

#Cloud SQL

#Google Pub/Sub

#Cloud Functions

#Google TTS

#Open AI

#RAG

#LangChain

#Terraform

Contextual Search

Contextual Search

Developed a cutting-edge Contextual Search Application catering to analysts' needs. Leveraging vector similarity search and Redisearch with RAG, the application enables analysts to input queries and efficiently retrieve top-k statements from earnings call transcripts by ensuring a user-friendly and intuitive interface for enhanced usability.

#Google compute engine

#Airflow

#Redisearch

#FastAPI

#Streamlit

#Docker

#Terraform

#OpenAI

#RAG

education

certifications

Deep Learning Basics

Deep Learning Basics

PadhAI - One Fourth Labs

Deep Learning | Pytorch | Image Classification Models

Certificate Credential