Software Engineer @ Ethara.ai
Building infrastructure for frontier AI systems.
LLM Evaluation • Agentic Coding • Benchmark Engineering • AI Infrastructure
|
Intelligent routing layer that directs queries to the optimal model by complexity, cutting inference cost and latency without sacrificing response quality. Python · FastAPI · React · TypeScript · Gemini API |
Full-stack image-generation platform serving prompt-driven synthesis through an interactive web app, with async inference and managed auth. React · Flask · Stable Diffusion · Hugging Face · Firebase |
|
Retrieval-augmented generation system pairing semantic search with LLM reasoning to answer questions over custom document collections. Python · LangChain · FAISS · OpenAI · Streamlit |
Real-time driver-monitoring system using computer vision to detect drowsiness, built on a low-latency YOLO inference pipeline. Python · YOLO · OpenCV · TensorFlow |
- Frontier LLM Evaluation — reproducible, sandboxed harnesses for scoring coding agents
- Agentic Coding Systems — pipelines that execute, verify, and self-correct model output
- AI Infrastructure — containerized, deterministic benchmark environments
- Benchmark Engineering — quality assurance and data-contamination detection at scale
- Open-source AI Tooling — developer tools for building and evaluating LLM systems
AI Infrastructure LLM Evaluation Agentic AI Developer Tools Benchmark Engineering Distributed AI Systems
I occasionally write about applied AI and engineering on Medium →
- Your Coding Agent Isn’t Solving the Bug. It’s Downloading the Answer.
- How Generative AI Is Changing Creative Work
- What Are Large Language Models (LLMs)?
