An LLM feature in your product: extraction, RAG or an agentJoe Estephan
An LLM feature that works on your data and ships with evals, so you know it works before your users do. Extraction from documents, retrieval over your content (RAG), or an agent that takes on a multi-step task.
This is my day job: I took an AI policy review service from an empty repo to production in seven weeks, then cut its cost per run by 68% with evals on every step. I also built the one-API LLM service behind it for Anthropic, OpenAI and Google models.
Tiers: Starter $500 (one feature over your data, 5 days). Standard $1,500 (a production service on AWS with tracing and cost tracking, 10 days). Advanced $4,000 (a multi-step agent or pipeline, 21 days).
To start: what the feature should do, a sample of the data it works on, and where it needs to run.
FAQs
Starting at$500
Duration1 week
Tags
Anthropic
AWS
OpenAI
Pinecone
Python
AI Engineer
Backend Engineer
Service provided by
Joe Estephan proVancouver, Canada
- 3
- Followers
An LLM feature in your product: extraction, RAG or an agentJoe Estephan
Starting at$500
Duration1 week
Tags
Anthropic
AWS
OpenAI
Pinecone
Python
AI Engineer
Backend Engineer
An LLM feature that works on your data and ships with evals, so you know it works before your users do. Extraction from documents, retrieval over your content (RAG), or an agent that takes on a multi-step task.
This is my day job: I took an AI policy review service from an empty repo to production in seven weeks, then cut its cost per run by 68% with evals on every step. I also built the one-API LLM service behind it for Anthropic, OpenAI and Google models.
Tiers: Starter $500 (one feature over your data, 5 days). Standard $1,500 (a production service on AWS with tracing and cost tracking, 10 days). Advanced $4,000 (a multi-step agent or pipeline, 21 days).
To start: what the feature should do, a sample of the data it works on, and where it needs to run.
FAQs
$500