Harsh Chaudhari

Software Engineer

Pune, Maharashtra, India4 yrs 8 mos experience

AI EnabledAI ML Practitioner

Key Highlights

Developed AI-powered systems with 99.9% uptime.
Led migration to PostgreSQL, enhancing performance.
Increased test automation coverage by 30%.

Stackforce AI infers this person is a Backend Developer specializing in SaaS and Fintech solutions.

Contact

f20190829@goa.bits-pilani.ac.in LinkedIn

Skills

Core Skills

Generative AiRetrieval-augmented Generation (rag)Node.jsPostgresqlTest Automation

Other Skills

FirebaseGoogle Pub/SubApache KafkaAlgorithmsHTMLPlaywrightJavaScriptBackend DeveloperAI AgentsLarge Language Model Operations (LLMOps)LanggraphLarge Language Models (LLM)Cloud FirestoreCloud functionsCursor

About

I’m a results-driven Software Developer with strong expertise in building scalable and resilient backend systems using Node.js, .NET (C#), and cloud-native technologies like Firebase, GCP Pub/Sub, Firestore, Redis, and Kafka. Currently at Fenado AI, I’m building an AI-powered RAG system that enables real-time querying over personalized Google Drive data using event-driven architecture and semantic search techniques. Previously at ICICI Lombard, I developed mission-critical backend services including a document management system that handled 50,000+ daily uploads, optimized Redis-based caching, and contributed to Elevate, a flagship insurance product serving 1L+ daily active users with ₹1200 Cr annual revenue. I also led the migration of legacy databases from Oracle to PostgreSQL, significantly improving performance. Earlier, during my internship at Blue Yonder, I enhanced automated testing coverage by 30%, helping deliver stable enterprise solutions to clients like IKEA, NESTLE, and NIKE. I enjoy solving complex backend problems, designing clean APIs, and implementing event-driven microservices architectures. 🛠️ Tech Skills: Node.js, C#, .NET Core, Firebase, Pub/Sub, Kafka, PostgreSQL, MongoDB, Redis, C/C++, REST APIs, SQL Server 📬 Let’s connect if you’re hiring for backend engineers passionate about performance, scale, and clean architecture! Contact @ +91 8459895279 Email @ harshchaudhari2001@gmail.com

Experience

4 yrs 8 mos

Total Experience

1 yr 3 mos

Average Tenure

8 mos

Current Experience

People interactive

Software Engineer - II

Oct 2025 – Present · 8 mos · Mumbai Metropolitan Region · Hybrid

Fenado ai

Software Developer

Apr 2025 – Oct 2025 · 6 mos · Noida, Uttar Pradesh, India · On-site

Built a fully automated RAG-based test generation pipeline integrated with GitHub via OAuth and commit webhooks, used LLM-driven prompt chains to parse README/docs, retrieve relevant context via semantic embeddings, and generate 50K+ AI-powered test cases/day, reducing manual QA creation time by 90%.
Implemented continuous semantic similarity retrieval using vector embeddings and cosine similarity over both retrieved and generated test cases, flagged duplicates with 95% precision, eliminating 40% redundant test specs via the RAG retriever.
Designed a full end-to-end QA platform underpinned by LLM + retrieval augmentation, includes bug management, test-runs tracking, GitHub activity dashboards, and generated test case modules delivering a 50% QA productivity boost, 3× improved coverage, and 75% fewer bugs in production.
Scaled platform adoption to 2,500+ teams using RAG-enhanced system, generating 50K tests/day via generative LLMs grounded in user repo context, maintaining 99.9% uptime SLA, and enabling 15hrs/week time saved per team.
Engineered a RAG-based financial knowledge platform integrating Google Drive for personalized document ingestion—processing 100% of uploaded files via webhook-triggered Firebase Cloud Functions; embedded and stored 50K+ document chunks in Firestore for sub-second retrieval.
Designed and deployed an event-driven ingestion pipeline using Firebase Cloud Functions and Google Pub/Sub, achieving end-to-end synchronization on file changes (adds/updates/deletes) with zero data loss.
Developed a semantic similarity retrieval engine using cosine similarity over embeddings, returning top-K context chunks in milliseconds—enabling 95% relevance accuracy and reducing LLM context size by 80%.

Retrieval-Augmented Generation (RAG)Generative AI

Icici lombard

Associate Software Engineer

Jun 2023 – Apr 2025 · 1 yr 10 mos · Mumbai, Maharashtra, India · On-site

Core contributor to Elevate, a high-revenue health insurance platform with 100K+ DAUs and 1200 Cr annual revenue.
Developed a Node.js-based document management system processing 50,000+ daily uploads with secure encryption, concurrency control, and robust error handling.
Built scalable backend logic for insurance workflows, significantly improving system responsiveness and reliability.
Implemented Kafka consumers and sync pipelines to maintain consistency between third-party data sources and internal systems.
Led the migration of legacy data and schemas from Oracle DB to PostgreSQL, optimizing queries for enhanced performance and maintainability.
Designed and implemented a Redis caching and data sync service using worker threads, reducing API latency by 30%.

PostgreSQLApache KafkaNode.js

Blue yonder

Software Engineer

Jul 2022 – Dec 2022 · 5 mos · Hyderabad, Telangana, India · Hybrid

Automated and expanded test coverage using Playwright (JavaScript), increasing test automation coverage by 30% and reducing post-release bugs by 25%.
Wrote and maintained 400+ scalable test scripts to validate critical production workflows.
Identified and resolved high-priority issues in enterprise platforms used by IKEA, NESTLE, and NIKE, directly contributing to delivery quality and customer satisfaction.

AlgorithmsHTML