Sai primarily worked as an Application Engineer building AI solutions at companies like Tekion, Paytm, TestMu & SigIQ.ai

github

Sai Ashish Darapureddy  • Peerlist

Razorpay Splash

Clients can add items to their cart and checkout to successfully place an order. Every nth order gets a coupon code for 10% discount and can apply to their cart.

Ecommerce Store

Responsive Angular PWA integrated with MongoDB Atlas database, Ngrx & Spring Boot backend.

Amazon PWA

MVP for Smart India Hackathon participation using Clarifai CNN deep learning AI models and Nutritionix.

NutriYou

LinkedIn GraphQL

This pattern was ideal for managing a multi-step loan application flow (like 💳 Loan Details → 🏦 Bank Account → 🔐 KYC), where each step independently validates and passes control to the next.

CRED's Flutter Chain of Responsibility

Engineered the data model using React, Docker, Webpack, MongoDB, SQL & Spring Boot

Sprinto Policies Platform

🚀 1,092 pattern-based problems in the revision sheet 💡 Java solutions for the first 500 questions ⚡️ Bonus PDFs covering System Design & Problem-Solving techniques

Leetcode2.0

Built using using React Native, Golang, Axios, PostgreSQL, Docker, Amazon RDS, Docker & Ngrok tunnel

InstaGo

Automated HLD Low Code tools via Make, WeWeb, Airtable, Xano DB & FlutterFlow.

Atlas Analytics Automation

NPMP

Full-stack app from scratch to help one decide whom to hire using the LinkedIn post reactions.

Mercor $100B Jobs

Built with k8s, FastAPI, & Huey, the startup was officially recognised by Google.

Camb.ai Multi Tenant Redis Store

Razorpay Google OAuth

Agentra is an educational project focused on understanding and building Agentic AI systems. AI agents that can reason, plan, use tools, and execute multi-step tasks autonomously.

Agentra

CRED's Splash Mock

Github

Aim of this project is to consume a short description to generate a short clip of a person speaking about the topic using GPT-3, GTTS and text-davinci-002 to handle the processing. This was my college major project, focused on generating a short video of a person speaking, based on a brief text description of the topic. It can also be leveraged to create advertisement clips based on a given topic. Objective: ✅ Generation of content from the topic or short description provided. ✅ Generation of an appropriate audio clip using the content from the previous step. ✅ Deepfaking a suitable video to sync the facial movement to the audio clip. Research & Methodology: ✅ We take short string of text as input. ✅ Process the text through OpenAI Generative Pretrained Transformer 3 (GPT-3) Model. ✅ Utilize text-davinci-002 to handle the processing. ✅ Utilize GTTS (Google Text-to-Speech) to handle text to speech processing. ✅ Display the results to user by using a React based Website. Limitations: ✅ Noisy and cluttered backgrounds in videos cause distortion artefacts in the final video. ✅ The resulting speech is not perfectly human in nature. ✅ Pronunciation of acronyms from written text is a concern.

Short DeepFake Video Generation from Text

Identification of profitable trading strategies, and monitor various kinds of technical indicators

Average-Directional-Index

Shortlisted for the Global Teams Challenge 2022, featuring several improvements across teams.