Remote (Anguilla, Antigua and Barbuda, Argentina, Aruba, Bahamas, Barbados, Belize, Bermuda, Bolivia, Bonaire, Sint Eustatius and Saba, Brazil, Cayman Islands, Chile, Colombia, Costa Rica, Cuba, Dominica, Dominican Republic, Ecuador, El Salvador, Falkland Islands (Malvinas), Grenada, Guatemala, Guyana, Haiti, Honduras, Jamaica, Mexico, Montserrat, Nicaragua, Panama, Paraguay, Peru, Puerto Rico, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, South Georgia and the South Sandwich Islands, Suriname, Trinidad and Tobago, Turks and Caicos Islands, Uruguay, Venezuela, Virgin Islands, British, Virgin Islands, U.S.)ContractEnglish JD· Posted Nov 1, 2026
From ITMates’s original posting
Client: mid-size US startup (Series B, about 150 people) · Location: remote, LATAM, 4+ hours overlap with US Eastern · Level: mid Engagement: full-time contractor
You will ship LLM-powered features into a B2B SaaS product, working with product engineers from prototype to production.
What you will do
- Build retrieval-augmented generation pipelines and agent workflows on top of commercial and open-source LLMs
- Write and version prompts, tools and evaluation sets
- Integrate AI features into the product's Python and TypeScript services
- Measure quality, latency and cost, and improve them release over release
- Add guardrails, logging and tracing for production use
What you bring
- 3+ years of software engineering, at least 1 building LLM applications in production
- Strong Python; experience with an LLM framework or SDK (LangChain, LlamaIndex or direct provider APIs)
- Vector databases (pgvector, Pinecone or Weaviate) and embedding-based search
- API design, testing and CI/CD habits
- Advanced English
Nice to have
- Evaluation tooling (LangSmith, Braintrust, Ragas)
- TypeScript and React
- Fine-tuning or classical ML background