RemoteAIFull-time
AI Engineer
Push our AI flagship Jimbo forward — agentic conversation, retrieval, and tool use across real customer channels.
- Location
- Remote (Nepal / GMT±5)
- Schedule
- 5 days / week · async-first
- Salary
- NPR 2.4M–4.2M / yr
- Level
- Mid–Senior
- Openings
- 1 seat
- Apply by
- July 31, 2026
What you'll do
- Design and ship LLM-powered agent loops that handle real conversations end to end.
- Build retrieval, grounding, and tool-use pipelines that stay accurate under load.
- Evaluate models and prompts rigorously — measure, don't guess.
- Take features from prototype to production with the rest of the product team.
What we're looking for
- 3+ years in software engineering, with hands-on LLM application work.
- Strong Python and/or TypeScript; comfortable across the backend stack.
- Practical experience with RAG, embeddings, and agent / tool-calling patterns.
- A healthy scepticism about model output and a habit of evaluating it.
Nice to have
- Experience with the Anthropic or OpenAI APIs in production.
- Familiarity with vector stores and streaming responses.
- Background in conversational or multi-channel systems.
What we offer
- Flexible hours and a remote-friendly culture
- A real say in what we build and how we build it
- Hardware and a learning budget that actually gets used
- Work that ships — and that you can point to
Stack
PythonTypeScriptLLMsRAGNode.jsPostgres
Ready to apply?
Send us your CV or a link to your work and a short note. It takes a few minutes.