Full-stack engineer who spent the last two years shipping production MERN systems and, over the last year, taught myself the AI/ML stack properly — LangChain, RAG, LLM agents, deep learning — rather than just reading about it. I specialize in Micro-Frontend architecture, scalable backends, and I like building things that actually get used: chatbots that stay grounded in real documents, agents that can query a database in plain English, backends that hold up under real traffic.
|
✅ Publicis Sapient — Frontend Engineer Client: Fidelity International · Gurgaon · Apr 2025 – Dec 2025
✅ Ijona Service LLP — Full Stack Engineer Bangalore · Feb 2024 – Aug 2025
|
|
🎨 Frontend Technologies
⚙️ Backend Technologies
🤖 AI / GenAI / ML
LangChain LCEL RAG AI Agents LangChain SQL Toolkit OpenAI API Hugging Face GROQ API LLaMA3 Ollama AstraDB TensorFlow/Keras ANN RNN LSTM GRU Transformers
🗄️ Databases & Caching
🔧 DevOps & Tools
|
📄 RAG Document Q&A Chatbot LangChain · GROQ API · LLaMA3 · AstraDB A retrieval-augmented pipeline over PDFs — chunking, embeddings, AstraDB vector storage — that keeps answers grounded, traceable to source, and aware of prior conversation.
|
💬 Conversational Q&A GenAI App LangChain · LCEL · OpenAI API · Streamlit A chatbot built on LCEL chains with a Streamlit front end that remembers the conversation across turns.
|
|
🗃️ Chat with SQL Database *LangChain SQL Toolkit · Agents · Python * A natural-language-to-SQL agent that turns plain-English questions into queries against a live relational database.
|
🔎 AI Agent + Search Engine Integration LangChain Tools & Agents · OpenAI API A tool-using agent that pulls in search and utility tools to work through multi-step queries that need outside lookups.
|
|
✍️ LSTM Next-Word Prediction TensorFlow/Keras · LSTM · NLP An LSTM language model trained from scratch — tokenization, sequence padding, embedding layers — to predict the next word in a sequence.
|
🧮 Additional ML/DL Work TensorFlow/Keras · LangChain ANN-based classification model (regularization, dropout, LR scheduling) plus LLM-based utilities for text summarization and math word-problem solving. |
|
🏢 Enterprise MFE Platform Publicis Sapient | Fidelity International 
Impact: Reduced production defects & improved maintainability |
🚗 WeMove — Ride, Bus & Hotel Booking Platform Node.js · Express · MongoDB · Redis · Socket.io 
|
|
🛢️ Oil Nomination System Sharjah National Oil Corporation (SNOC)
Impact: Government-grade oil nomination system |
🏃 Sport & Health App React Native Fitness Platform 
|
| 🎯 Metric | 📈 Achievement |
|---|---|
| Performance Boost | ⚡ 30–40% faster API responses |
| Code Quality | ✅ 90% test coverage |
| Migration | 🔄 Legacy Backbone.js → Modern React |
| Security | 🔐 JWT/OAuth2 & RBAC expert |
| Real-time Systems | 📡 WebSocket implementations |
| Architecture | 🏗️ Micro-Frontend specialist |
| AI/GenAI | 🤖 RAG pipelines, LLM agents, LangChain |
| Scale | 👥 Apps serving 5,000+ daily active users |
|
|
Certifications: Complete Generative AI Course (Tutedude) · AI Fundamentals (IBM) · Python Programming & Backend Development (Tutedude) · TypeScript Certification (Publicis Sapient) · Full Stack Web Development (Masai School, Bangalore)
- ⚡ Performance optimization enthusiast — consistently achieving 30–40% improvements
- 🌍 International project experience across India, Dubai, and South Africa
- 🧩 Passionate about clean code architecture and best practices
- 📚 Pursuing an MCA while working full-time as a developer
- 🚀 Actively exploring Micro-Frontend architectures and the GenAI stack
- 🔥 70+ public repositories, 2,500+ commits and counting
AI/GenAI Engineer | Full Stack Developer | Frontend Engineer | Freelance Projects
RAG & LLM Agents | Micro-Frontend | Enterprise Applications | Performance Optimization | Scalable Systems
📧 arunkumar08.mk@gmail.com | 📞 +91 9718653508
⭐ From Arun Jawlia
Thanks for visiting! Let's connect and build something amazing together! 🚀