Writing
Platform comparisons, latency engineering, real costs, and the parts of a voice pipeline nobody warns you about.
Four ways to build a voice agent, compared on latency, real cost per minute, control and the work you inherit. Written by someone who has shipped on more than one of them.
Read the articleBuild costs from $4,000 to $45,000, running costs from $0.04 to $0.25 per minute. A line-by-line breakdown of where the money actually goes, and the costs that never appear in the first quote.
Read the articleUnder 800ms feels alive. Past 1.5s people talk over the agent. A stage-by-stage breakdown of the latency budget and the eight fixes that actually move the number.
Read the articleSpeech-to-speech wins on latency and naturalness. Chained pipelines win on control, cost and auditability. How to pick, and why most production systems I build end up chained.
Read the articleThe four reasons that justify a migration, the work managed platforms were quietly doing for you, and a cutover sequence that does not let customers discover the new stack is worse.
Read the articleLeaked model API keys are one of the most common and most expensive AI production failures. Server-side proxying, key scoping, budget caps and the monitoring that catches abuse early.
Read the articleApp development cost ranges from $1,200 to $200,000+ depending on complexity, platform, and team. Realistic 2026 pricing, the 7 factors that drive cost, what AI actually saves you, and the hidden fees no one puts in the first quote.
Read the articleDirect Project Enquiry
Share what you are building and where it is stuck. If a managed platform already solves it, I will say so. I review every enquiry personally and reply within about four hours.
International projects welcome
I work in English by email, video call or through Upwork, from GMT+7 with overlap hours for US and EU teams.
Read the Privacy Policy, Terms of Service and Cookie Policy.