Hi, I'm Hariom Acharya.
I care more about shipping something real than talking about how I'd build it. Most of what I know came from taking a project past the point where it just works on my machine — into something a real user could open, break, and depend on. That bias toward building shows up in everything I do: I'd rather have a rough version running end-to-end by day two than a perfect plan on paper by day five.
My core strength is applied AI — not just calling an API, but designing prompts and systems that hold up outside a demo: reliable outputs, sensible fallbacks, and integrations that fit into a real product rather than sitting next to it. Alongside that, I work full-stack — React and TypeScript on the frontend, Node.js/Express and Supabase on the backend — and I'm actively building out my cloud fundamentals: AWS, Docker, and the deployment patterns that turn a working app into a production one. I also contribute to open source when I can, because reading and improving someone else's codebase teaches discipline that solo projects never do — you write code knowing someone else has to review it.
Skills & Stack
5 focus areasFrontend & UI Engineering
Architecting performant, accessible web interfaces and component systems with React, TypeScript, and Tailwind CSS. Obsessed with 60fps micro-animations, fluid layout transitions, and intuitive design ergonomics that elevate user engagement.
Backend & Systems Design
Engineering production-grade REST APIs, relational schemas, and server logic with Node.js, Express, and PostgreSQL/Supabase. Focused on secure token authentication, low-latency queries, and strict data consistency under load.
Applied AI & Intelligence
Designing high-leverage LLM pipelines, prompt architectures, and custom classification models with Python and OpenAI. Bridging the gap between raw models and practical applications through automated inference and semantic parsing.
Cloud Infrastructure & DevOps
Containerizing distributed applications with Docker, provisioning scalable AWS serverless infrastructure, and architecting automated CI/CD pipelines to ensure rapid, zero-downtime deployments with observable health checks.
IoT & Edge Telemetry
Interfacing physical microcontrollers (ESP32) and industrial sensor arrays over real-time protocols (MQTT/HTTP). Streaming real-world vibration, temperature, and electrical telemetry into predictive edge models to forecast remaining useful life.
Where My Time Goes
My hands-on work spans intelligent platforms like Vox-Hire, an AI-powered mock interview system, and PhishGuard, a real-time phishing detection system combining threat feeds with custom classifiers. I'm also drawn to hardware — I built an ESP32-based Machine Failure Detection pipeline streaming live telemetry into ML models to estimate remaining useful life.
I approach all of it with a structured, SDLC-driven mindset: understand the problem thoroughly, plan before building, test rigorously rather than ship and hope. Right now I'm wrapping up my final year while sharpening cloud fundamentals — AWS, Docker, serverless patterns, and tighter AI integration across my projects.
I'd rather ship something small and working than plan something large and theoretical.