AI Engineer building production-ready AI systems, agentic workflows, and edge intelligence.
I enjoy turning research and ideas into reliable software—from LLM-powered applications and voice agents to computer vision and real-time inference. Most of my work focuses on designing systems that are practical, scalable, and easy to deploy.
- 🤖 Production AI systems & Agentic AI
- 🎙️ Speech AI (ASR, TTS, Voice Agents)
- 👁️ Computer Vision & Edge AI
- ⚡ MLOps, GPU Inference & AI Deployment
- 💬 Gesprek — Multi-agent voice platform with real-time conversations
- 💰 Hishab — AI-powered expense tracker using agentic workflows
- 📷 StereoLite — Lightweight stereo depth estimation for edge devices
- 🥗 Diet Engine — AI nutrition assistant with real-time food recognition
Python • Go • FastAPI • Django • Flutter
PyTorch • TensorFlow • Transformers • LangGraph • Hugging Face
PostgreSQL • Redis • Docker • Supabase • LiveKit • WebRTC
YOLO • ONNX • TFLite • MLX • AWS • Linux
LinkedIn • Kaggle • Codeforces • Email
I like building AI systems that solve real problems—not just demos.



