Aishik Dalui

About Aishik Dalui

I’m a passionate React Native Developer with 2 years of hands-on experience building high-performance, scalable mobile applications. My journey started with developing intuitive, efficient, and optimized mobile experiences — from improving app load times and creating seamless UIs to integrating complex APIs and managing global state with Redux Toolkit and Redux-Saga.Recently, I’ve been expanding my skill set into the AI and LLM ecosystem, exploring how modern technologies like Tokenization, Agents, Retrieval-Augmented Generation (RAG), Vector Databases, and AI app deployment can be leveraged to build intelligent, production-grade systems. I enjoy connecting mobile and AI worlds — creating apps that are not only functional but also context-aware and smart.I’m always driven by learning, problem-solving, and building solutions that scale — whether it’s optimizing a React Native app’s performance or deploying AI-powered agents in production.🧠 Core Skills:React Native Development (Redux Toolkit, Redux-Saga, MMKV)API Integration & OptimizationTokenization & AgentsRetrieval-Augmented Generation (RAG)Vector Databases (e.g, Pinecone, FAISS, Weaviate)Deploying Scalable AI ApplicationsEnd-to-End Mobile + AI Integration

Experience

1. Software Engineer

   Mantra Labs

   May 2024 — Present · Kolkata, IN

   Contributed to Connect2Clinic, an OPD practice management platform enabling in‑clinic and tele‑consult scheduling, secure video calls, smart eRx, and doctor/patient health‑record uploads to streamline end‑to‑end outpatient workflows. Delivered features spanning payments, queue management, dynamic walk‑ins, WhatsApp booking, automated notifications, and appointment insights to improve operational efficiency and patient experience. — Implement MMKV‑backed login and two‑level clinic/API caching to persist default clinic context and reduce redundant calls, cutting load time from 10–15s to 5–7s while enhancing User Experience and conserving battery life. — Integrated debounced medicine search and FlatList lazy loading/virtualization to minimize redundant API calls and render only visible items for large inventories, improving responsiveness and scalability. – Optimized rendering by leveraging React.memo with useMemo/useCallback to prevent unnecessary re-renders, and built cross‑platform downloads with rn‑fetch‑blob while Collaborate Amazon S3 for secure file upload and in‑app viewing.

Education

• Narula Institute Of Technology

  B.Tech, Electronics and communications Engineering

  2019 — 2023

• Narula Institute of Technology

  Bachelor of Technology, Electrical, Electronics and Communications Engineering

  2019 — 2023