AI Engineer · Bangkok

I build AI features that work in production.

I help teams turn AI ideas into reliable products: assistants grounded in your own data, agents that automate real workflows, and evals that tell you whether any of it works.

$ building

Services

From “could AI help here?” to a shipped feature

LLM-powered apps

Chat assistants, copilots and AI features built into your product, with good prompts, structured outputs and a UX people actually use.

RAG & knowledge systems

Let AI answer from your documents, help center or database, with accurate retrieval and answers that cite their sources.

AI agents & automation

Agents that call your tools and APIs to handle multi-step work like triaging tickets, processing documents or drafting reports.

Evals & reliability

Test suites, guardrails and monitoring so you know when the AI is right and catch regressions before your customers do.

AI integration

Add AI to what you already have. I work inside existing codebases and connect models to your CRM, CMS or internal tools.

Full-stack delivery

From prototype to production: frontend, backend, deployment and cost control, without handing off between several contractors.

Tools I build with

  • Claude
  • OpenAI
  • Python
  • TypeScript
  • Next.js
  • React
  • LangGraph
  • MCP
  • pgvector
  • Pinecone
  • Postgres
  • Vercel

Also powered byPizzaCatsBooksYoga

Lise Carpenter

About

Hi, I'm Lise

I'm an AI engineer with a full-stack web background and over a decade in financial services and tech startups. I've spent my career shipping software for real users, and now I bring that experience to AI: I build the retrieval pipelines, agents and evals that turn a promising demo into something your business can depend on.

Coming from fintech, I care about accuracy, privacy and cost as much as the model itself. I'll tell you plainly when AI is the right tool and when a simpler solution will do.

Working with me

I'm based in Bangkok, work with clients worldwide, and speak English, Thai and French. Engagements usually start with a short scoping call, then a focused prototype, so you can see real results before committing to a full build.

Contact

Have an AI idea or problem?

Tell me what you're trying to do. I'll reply within two business days, and the first call is free.