Who I am

I build production software with AI woven through the entire stack, from the code itself to the product features running on top of it. My core focus is applied AI: I architect multi-provider LLM integrations behind provider-agnostic routing, design agentic development workflows where AI coding agents handle real engineering work under human review, deploy open-weight models locally for private inference, build RAG systems over proprietary data, and fine-tune open-source models for domain-specific tasks. I've also done hands-on model evaluation work with Outlier by Scale AI, assessing LLM outputs for bias, accuracy, and multilingual quality.

That technical side sits on top of 8+ years leading customer support, CX, support operations, and Trust & Safety teams in high-performance environments, most recently as a Subject Matter Expert at TDCX supporting Meta's WhatsApp and Central Integrity projects. In that role I act as the bridge between agents, QA, and management: running policy calibration sessions with clients, leading root cause analysis on complex escalations, and driving training rollouts for new knowledge and regulatory updates. I've mentored and coached teams to help new hires ramp up quickly, and I bring hands-on CX experience from managing live chat, email, inbound, and outbound operations at scale on the Traveloka account with ATI Business Group.

I use tools like Salesforce, SQL, Power BI, and Tableau to monitor performance and turn operational data into decisions, and I lean on that same data-driven instinct when I'm architecting AI systems. That combination gives me a genuinely rare angle: I don't just build AI products, I understand how they hold up against real abuse, ambiguity, and scale, because I've spent years on the front line of exactly that problem.

Open to conversations on AI engineering, agentic systems, and trust & safety operations.