Portfolio
AI/ML Engineer & Full-Stack Developer
Full-stack engineer who ships production LLM/RAG applications as readily as enterprise .NET systems — equally at home as an AI/ML Engineer or a Full-Stack Developer.
Scroll
About
I'm Ravi Kiran — a Software Engineer who qualifies equally as an AI/ML Engineer and a Full-Stack Developer, with 5+ years in the Pharmacy Ecosystem. My signature is the unglamorous engineering work that keeps healthcare-grade systems running: a codebase-wide Timezone Migration with NodaTime/IANA spanning 300+ screens, hardening Auth0 with WebAuthn passkeys and MFA, fixing SQL Server connection leaks in non-idempotent retry pipelines, and standing up EDI transaction processing for billing and eligibility.
Beyond that I've shipped event-driven Kafka pipelines, Eyecon integration for pill verification, 340B compliance, Compounding Rx workflows, and a Patient Engagement platform built on PWA architecture — all with AI tools (Claude Code, Cursor, Copilot) woven into the daily loop.
On the Generative AI side I build end-to-end LLM applications across all three major model providers: PortfolioBot — the chat assistant on this page — is a production RAG system I designed from scratch (FastAPI, OpenAI GPT-4o-mini, text-embedding-3-small, ChromaDB) with a custom Router that classifies Personal vs General questions and returns cited sources. Remindly is a Flutter reminder app with Google Gemini for natural-language scheduling and photo-to-reminder. ScrumAgent is an agentic AI assistant on Anthropic Claude that turns meeting transcripts into verified Jira tickets via a tool-use loop. I'm hands-on with Retrieval-Augmented Generation, Agentic AI, Vector Databases (ChromaDB, Pinecone), Embeddings & Semantic Search, Prompt Engineering, and OpenAI / Anthropic / Gemini API integration across FastAPI & Python backends.
I'm open to AI/ML Engineer, Full-Stack Developer, and Software Engineer roles — see my work on GitHub and LinkedIn.
Signature Work
Specific, measurable engineering wins shipped in production pharmacy systems.
Led UTC storage with NodaTime/IANA, SQL Server AT TIME ZONE, and Angular Localization across the entire app. Interceptor-based and service-based approaches.
Refactored prescription lookup queries — reduced response times across the busiest pharmacy screens.
Diagnosed and rewrote retry-pipeline scoping and async patterns that were exhausting the connection pool under load.
WebAuthn/Passkey biometric flows, Phone MFA, Email OTP, and SSO certificate management for secure patient authentication.
Built EDI flows for pharmacy billing, insurance claims, and real-time eligibility verification.
Apache Kafka pipelines for real-time pharmacy data streaming and transaction messaging across services.
Tech Stack
Technologies I use daily to build reliable, scalable software.
Experience
From intern to engineer — building healthcare software pharmacies depend on daily.
Projects
Production RAG assistants, enterprise pharmacy platforms, and AI applications.
Cloud Modernization across PMS, Wholesale Market Integration, POS for Checkout, and Patient App — centralized Prescription Processing, distributed Elasticsearch Search, real-time Transaction Handling, 340B Compliance, Compounding Rx workflows, and cross-location synchronization over legacy on-prem systems.
The chat widget on this very page — a Retrieval-Augmented Generation (RAG) assistant grounded in my own resume, experience, and projects. Built end-to-end: FastAPI backend on OpenAI GPT-4o-mini, text-embedding-3-small for semantic search, ChromaDB persistent vector store, and a custom Router that classifies questions as Personal (RAG-grounded with cited sources) vs General (direct LLM). Embeddable vanilla-JS widget on the frontend, CORS-locked, deployed on Render.
Patient-facing pharmacy platform on PWA Architecture — Biometric Login (Face ID), Rx Refills, Rx History, Insurance Management, Due-Date Tracking, Pharmacy Chat, Medication Profiles. Appless install, no app store needed.
Total Pharmacy Dispensing — Claims, Labeling, Inventory Management, Rx Generation.
Flutter (Android/iOS) reminder & to-do app — full-screen alarm-style reminders that keep nudging until done. Gemini-powered natural-language add ("charge toothbrush at 9pm"), photo→reminder, AI list generation, assistant chat, and weekly insights. Offline-first, Firebase sync, AI degrades gracefully.
Agentic AI assistant for Scrum Masters, built on Anthropic Claude with a tool-use loop. Turns meeting transcripts into Jira tickets, verifies its own output against the source to prevent hallucination, drafts follow-up emails, and writes sprint health reports. 45 tests, production-hardened retries/backoff.
Built in Angular from scratch — OpenAI / Anthropic API integration for Dynamic Question Generation, Real-Time Scoring, animated Leaderboards. 300+ live participants.
Voice-Command Email System — Compose, Check, Search, and Delete emails via voice instructions using Python.
Real-Time CNN Face Mask Detection using OpenCV for public safety — airports, offices, schools.
Raspberry Pi + OpenCV Real-Time Drowsiness Detection with automatic alarm alerts for long-distance travel safety.
Education
Strong foundations from two continents.
Montclair State University
2024 Jan – 2025 Dec
IIIT Nuzvid (RGUKT), AP, India
2017 – 2021
Sails Software Solutions — Best Performer for MMS Project, Real-Time Data Solutions & Production Support.
Sails Software Solutions — Outstanding UI Development and Code Refactoring.
Andhra Pradesh Government — State-Level Academic Excellence Recognition.
Contact
Always open to new opportunities and interesting projects.
Software Engineer qualified for both AI/ML Engineer and Full-Stack Developer roles, with 5+ years in the Pharmacy Ecosystem — shipping production software across PMS, Wholesale Market Integration, POS for Checkout, and Patient App. Deep domain expertise across Prescription Processing, eRx Automation, Inventory Management, Drug Price Updates, EDI Transactions, Insurance Eligibility, Delivery and Shipping, Tenant Migration, 340B Compliance, Compounding Rx, and Patient Engagement on PWA Architecture. Strong full-stack foundations in Angular, .NET (C#, ASP.NET Core), SQL Server, Elasticsearch, Kafka, Azure, Auth0/WebAuthn, paired with daily AI-augmented engineering using Claude Code, ChatGPT, Cursor, GitHub Copilot, Google Antigravity. AI / GenAI focus areas: Generative AI, LLM Application Development, Agentic AI, Retrieval-Augmented Generation (RAG), Vector Databases (ChromaDB, Pinecone), Embeddings & Semantic Search, Prompt Engineering, OpenAI / Anthropic / Gemini API Integration, and FastAPI / Python AI Backends. Built and deployed PortfolioBot — a production RAG assistant (FastAPI + OpenAI GPT-4o-mini + ChromaDB) embedded live on this site — Remindly, a Flutter reminder app with Google Gemini for natural-language scheduling and photo-to-reminder, and ScrumAgent, an agentic AI assistant on Anthropic Claude that turns meeting transcripts into verified Jira tickets. Platform-agnostic across OpenAI, Gemini, and Anthropic — I ship the AI feature, not just call the API. GitHub: github.com/pilliravikiran · LinkedIn: linkedin.com/in/ravikiranpilli
Dear Hiring Manager,
I'm writing to express my interest in an AI/ML Engineer or Full-Stack Developer position. With 5+ years in the Pharmacy Ecosystem — shipping production code across PMS, Wholesale Market Integration, POS for Checkout, and Patient App using Angular, .NET, SQL Server, Azure, and AI — combined with a Master's in Computer Science (3.9 GPA) and a portfolio of production LLM/RAG applications, I bring deep healthcare-domain expertise, modern cloud architecture skills, and hands-on Generative AI engineering experience that qualify me for either track.
I started in this ecosystem as a Software Engineer Intern at Sails Software Solutions in 2019, then grew into a full-time .NET Full Stack Developer role there building a PMS — pharmacy delivery, eRx automation, Kafka event-driven flows, and CI/CD on Azure DevOps. After completing my Master's, I returned to the same Pharmacy Ecosystem directly at RedSail Technologies, where I'm now building Patient Engagement on a PWA Architecture — appless, install-from-browser, no app-store distribution required — alongside 340B Compliance and Compounding Rx workflows for covered entities.
At RedSail, I've led initiatives impacting product quality and scale: timezone migration spanning 300+ screens, secure patient authentication with Auth0 WebAuthn biometrics and MFA, EDI transaction processing for billing and eligibility, and Elasticsearch optimizations improving prescription lookup by 30%. I've also fixed critical infrastructure issues like SQL Server connection leaks and non-idempotent retry pipelines.
On the Generative AI side, I design and ship end-to-end LLM applications. My production project PortfolioBot — the chat assistant running live on my portfolio site — is a Retrieval-Augmented Generation (RAG) system I built from scratch: a FastAPI backend on OpenAI GPT-4o-mini with text-embedding-3-small for semantic search, ChromaDB as a persistent vector store, and a custom Router LLM call that classifies questions as Personal (retrieves Top-K chunks and answers with cited source filenames) vs General (direct LLM). I authored the full ingestion pipeline (chunking, embedding, indexing), a three-role Prompt Engineering setup, and deployed the API on Render with CORS-locked access and an embeddable vanilla-JS widget on the frontend.
I've deliberately built across all three major model providers so I can pick the right tool rather than being locked into one API. Beyond PortfolioBot, I built Remindly, a Flutter reminder app where Google Gemini powers natural-language scheduling ("charge toothbrush at 9pm"), photo-to-reminder, and AI-generated to-do lists — with every AI feature degrading gracefully so the app is fully usable offline. I also built ScrumAgent, an agentic assistant on Anthropic's Claude that turns messy meeting transcripts into Jira tickets through a tool-use loop, then runs an independent verification pass so it doesn't hallucinate action items — backed by 45 tests and production-grade retry/backoff handling.
My broader AI / GenAI focus areas include: Generative AI, LLM Application Development, Agentic AI / Tool-Use Loops, Retrieval-Augmented Generation (RAG), Vector Databases (ChromaDB, Pinecone), Embeddings & Semantic Search, Prompt Engineering, OpenAI / Anthropic / Gemini API integration, LangChain, and FastAPI / Python AI Backends. I also use Claude Code, Cursor, ChatGPT, GitHub Copilot, and Google Antigravity daily to accelerate feature work, refactors, and reviews. Earlier, my AI-Powered Quiz Platform — built in Angular with OpenAI and Anthropic APIs for dynamic question generation and real-time scoring — served 300+ live participants.
I'm drawn to teams that value engineering rigor, thoughtful architecture, and continuous innovation — and I'd welcome the chance to bring this combination of pharmacy-domain depth, full-stack execution, and production GenAI engineering to your team.
Thank you for your time and consideration.