About
I am William Balance, a full-stack developer focused on applied AI. I design and ship web products that use LLMs to actually do something useful, not as a marketing sticker.
I work remotely from South Korea with clients in Europe, North America, the Middle East and Asia. I cover EU hours in the morning and US or APAC hours in the evening.
Main stack: TypeScript, Next.js, Astro, Node, Python, Postgres, Cloudflare, OpenAI, Anthropic, pgvector, Qdrant. I pick the tool that fits the problem, never the other way around.
How I work: clear scoping, short delivery cycles, readable code, costs under control, docs that actually get used. I would rather ship a feature that works in two weeks than a perfect system that never lands.
Frequently asked questions
Who is William Balance?
Freelance full-stack developer specialized in applied AI. I ship web products that use LLMs to actually do the job: internal assistants, domain chatbots, automations, AI features inside existing SaaS. Based in South Korea, I work with clients worldwide.
Why South Korea?
It is a deliberate choice. The tech ecosystem is serious, the infrastructure is flawless, and the time zone lets me cover Europe in the morning and the Americas or APAC in the evening. For clients, that means work happens while they sleep.
What is your experience with AI?
I have been shipping LLM integrations in production since GPT-3.5 landed. Day-to-day: Claude and OpenAI in production, RAG with pgvector and Qdrant, agents with tool use, MCP, multi-agent orchestration, light fine-tuning. Not theory — code running with real users.
What languages do you work in?
French and English fluently, Korean in progress. I take projects in FR and EN without friction, both in writing and on calls. Technical documentation is always in English to stay portable across teams.
How can I reach you?
Easiest path: book a free 30-minute slot via the link at the top of the page. Otherwise email me at [email protected]. I reply within one business day, usually faster.