AI & LLM Integrations

What we build

Meeting pipelines, transcription and diarization, embedding search over your knowledge base, agent tools that call your real APIs, voice front-ends and safety layers you can actually explain to enterprise buyers.

Where it shows up

The AI meeting-intelligence platform for the US startup and multiple internal automation projects across our client base.

Example

// Simple RAG query with pgvector
const embedding = await openai.embeddings.create({ model: "text-embedding-3-small", input: q });
const rows = await sql`
  SELECT content FROM docs
  ORDER BY embedding <-> ${embedding.data[0].embedding}
  LIMIT 5
`;
const answer = await openai.chat.completions.create({
  model: "gpt-4.1-mini",
  messages: [{ role: "system", content: "Answer only from context." },
             { role: "user", content: `Q: ${q}\n\nContext:\n${rows.map(r => r.content).join("\n---\n")}` }]
});