I'm a Machine Learning Engineer with 4+ years of experience turning "wouldn't it be cool if AI could..." into actual systems that run in production — and don't catch fire (usually).
I've built RAG pipelines that retrieve the right thing, inventory agents that know when you're out of guacamole, and InterVue Labs — an AI interview simulator that's been known to make humans question their own qualifications.
My superpower? Closing the gap between "here's a cool paper" and "here's a system people are actually using."
| 🎙️ |
InterVue Labs "What if your interview prep actually talked back?" I built InterVue Labs from scratch — An AI-powered interview simulator that conducts real, dynamic two-way conversations using LLMs, RAG, persistent memory, and voice interaction. Not a chatbot with a list of questions. An actual interview experience that adapts to your answers, pushes back when you're vague, and remembers what you said three questions ago. Because the best way to get good at interviews is to practice with something that doesn't go easy on you. Stack: Python · LangChain · RAG · Vector DB · Voice AI · FastAPI |
🪑🤖💬 InterVue Labs | Where AI puts you in the hot seat
- Advanced agentic patterns (multi-agent orchestration, tool use, reflection loops)
- Building Scalable Production Grade AI/ML Systems
Fellow developers and engineers who want to get in on something real.
- I'm actively looking for collaborators to help push InterVue Labs to the next level. Whether you're strong in backend architecture, LLM fine-tuning, voice AI, or just have opinions about what makes an interview tool actually useful — I want to hear from you. Fresh perspectives and brutal honesty welcome. The AI already judges candidates enough; it can handle your feedback too.
I once trained a model to detect personality traits from handwriting samples. It classified mine as "high attention to detail and analytical thinking." I had 47 typos in the paper. The model and I have agreed not to discuss it further.
