I build reliable browser systems that let AI agents access, navigate, and act on the web.
Browser InfrastructureAI AgentsApplied Reverse EngineeringML Systems
- Browser infrastructure: I build and harden scalable cloud and headless browser systems across authentication, state, identity, network behavior, isolation, observability, and failure recovery.
- Agent execution: I develop runtimes, sandboxes, browser skills, developer APIs and SDKs, and Model Context Protocol (MCP) tools. I evaluate deterministic automation and LLM-driven workflows for task success, latency, cost, and reliability.
- Browser internals and applied reverse engineering: I work across browser, renderer, JavaScript, graphics, networking, TLS, PDF, and DevTools layers. I also analyze undocumented web and mobile protocols and turn the findings into production systems.
- Applied AI and data systems: I train and evaluate machine learning models, fine-tune large language models (LLMs), and build data-intensive systems for research and production workloads.
Languages:Go · TypeScript · Python · C++
Browser and automation:Chromium Internals · Blink · V8 · Skia ·
PDFium · BoringSSL · CDP · Playwright · Puppeteer ·
Browser Automation · Web Scraping
AI and data:Machine Learning · Model Training · LLM Fine-Tuning ·
AI Agents · Computer Use · Agent Evaluation · Data Pipelines
Systems:Distributed Systems · Linux · Docker · Observability
Most of my recent work is private. I can discuss the problems, my role, and the outcomes. I keep client names and proprietary implementation details confidential.
Browser infrastructure · Agent platforms · Applied AI systems
hackme.oku2l@simplelogin.com