Hi, I'm Vivek
AI Engineer · Full-Stack Developer · Founder of OpnCrafter
My Story
I started seriously working with LLMs in early 2023, when the API for GPT-3.5 first became widely available. I was immediately hooked — not by the chatbot demos, but by the potential to build autonomous systems that could reason, plan, and act.
Over the next year, I built and shipped several AI-powered products: a voice-driven scheduling assistant, a RAG pipeline over legal documents, and a multi-agent research swarm. I made a lot of mistakes along the way — burning through API credits on bad prompts, debugging hallucinations at 2am, and fighting with JSON parsing errors from LLM outputs.
I started OpnCrafter because I couldn't find guides that were both technically deep and practical. Most tutorials show a happy path. I wanted to write guides that show the real path — including the dead ends.
What I've Built
RAG Pipelines
Hybrid search with Chroma, Pinecone, and Weaviate. Learned chunking strategies the hard way.
Autonomous Agents
LangGraph multi-agent systems for research, coding, and customer support.
Voice AI
Real-time voice bots with Deepgram and OpenAI Whisper, live demo deployments included.
Generative UI
RSC + Vercel AI SDK streaming components — the future of AI interfaces.
My Stack
Every guide I write is backed by real, tested code. Here's what I work with day-to-day:
What OpnCrafter Stands For
No Fluff
I skip the motivational intros. Every guide gets straight to working code and the reasoning behind it.
Open Source First
I champion open frameworks: LangChain, LlamaIndex, Ollama. AI should not be locked behind vendor walls.
Engineering Rigor
AI is mostly math and systems engineering. I focus on evaluations, tracing, and reliability over hype.
Let's Connect
Whether you're debugging an agent, exploring a new framework, or just want to talk AI — I'm reachable.
Get in Touch →