Fullstack Engineer

Prompting Company; San Francisco (CA); On-site; USD 80,000 - 150,000; Full time

Job summary

Prompting Company in San Francisco is seeking a founding engineer to design experiments and systems that reveal how AI agents crawl, retrieve, and cite the web. You’ll work on infrastructure with TypeScript, Next.js, and Go, build telemetry pipelines, and help products get discovered by AI agents for enterprise partners.

This role offers a collaborative, founding-team environment and the chance to shape how AI agents interact with web content.

Qualifications

• Experience with LLMs at infra or evaluation level.
• Built edge systems for scalable web apps.
• Shipped production systems in Next.js, TypeScript, or Go.
• Built or debugged telemetry and routing pipelines.
• Curious about how LLMs choose what to see, cite, and trust.

Responsibilities

• Design experiments and systems for AI agents crawling and citing the web.
• Run large-scale prompt experiments and agent workflow evaluations.
• Build infrastructure to simulate, trace, and debug agent behavior across docs, websites, APIs, SDKs, and product workflows.
• Track token-level cost, caching, deduplication, and latency for high-volume AI and agent traffic.
• Build analytics for share of voice, competitive visibility, workflow success rates, and agent failure points across models.

Skills

LLM infrastructure

Edge systems design

Next.js knowledge

Go backend experience

Telemetry & routing pipelines

Tools

Next.js

TypeScript

Go

Postgres

pgvector

Job description

Overview

We help products get discovered and used by AI agents.

Location & Compensation

San Francisco, CA, US – Full-time – US citizen/visa only – 1+ years – Salary $80K - $150K.

About the Role

We help companies get mentioned in ChatGPT by rebuilding the internet for AI agents. Our product helps companies like NVIDIA, Rho, and Exa get cited by ChatGPT, Gemini, and Perplexity.

You’ll join as a founding engineer to design experiments and systems that uncover how AI agents crawl, retrieve, and cite the web.

Who You Are

• Worked with LLMs at the infrastructure or evaluation level (OpenAI, Anthropic, Perplexity, LangChain, etc.)
• Built edge systems at Cloudflare, Vercel, or Netlify and want to apply that expertise to AI agents
• Shipped production systems in Next.js, TypeScript, or Go
• Built or debugged telemetry and routing pipelines
• Are curious about how LLMs choose what to see, cite, and trust

Technology & Challenges

• Crawl, transform, and serve structured content that LLMs and agents can understand, retrieve, and use.
• Run large-scale prompt experiments and agent workflow evaluations to understand how AI systems discover products, compare options, and complete tasks.
• Build infrastructure to simulate, trace, and debug agent behavior across docs, websites, APIs, SDKs, and product workflows.
• Track token-level cost, caching, deduplication, and latency for high-volume AI and agent traffic.
• Build analytics for share of voice, competitive visibility, workflow success rates, and agent failure points across models.

Our stack:

• Frontend: TypeScript, Next.js.
• Backend: Go.
• Data: Postgres and pgvector, Tinybird for analytics.
• Infra: Kubernetes, Cloudflare Workers, CDN-first edge routing.