Job description

Jan is a productivity company. We build a cross-platform, local-first and AI native framework that can be used to build anything. This includes https://jan.ai/ a desktop app that runs AI on your own laptop, 100% offline and privately. We support most popular AI models, and are actively working on a roadmap to allow users to customize and fine-tune these AIs to meet their specific needs.

We are a fully remote, open source company. We target the global market, but operate as a lean, bootstrapped company.

Job Description

Jan is looking for a LLM Engineer/Researcher to continue to fine-tune and train our own LLM models.

Responsibilities

Train and Fine-tune foundational LLM models (e.g. PEFT, Lora, QLora, latest research techniques)
Build and maintain LLM applications and infrastructure to meet business needs
Design LLM inference infrastructure to scalably deploy LLMs within infrastructural constraints
Research and utilize best of class tools within LLM ecosystem (e.g. Vector databases, LlamaIndex, etc)
Keep up with latest research around LLMs (e.g. sparse models, hardware-specific LLMs)
Research and keep up with latest use-cases of LLMs (e.g. RAG, Agents, etc)
Collaborate closely with LLM research teams to participate in foundation model research, specifically for training productivity-related LLMs.

Requirements

Experience with LLMs, including popular foundation models like Llama2, MPT
Experience with Training and Fine-tuning foundational LLM models
Experience with quantization techniques, including llama.cpp, GPTQ, etc
Experience with LLM related development, e.g. Llamaindex, Langchain, Vector DBs, Prompt Engineering etc
[Plus but not required] Experience running LLMs in production (e.g. Triton Inference Server, etc)

Benefits

We pay an “all-in” pay and you will cover your own insurance/medical from the amount
14 days leave (and unlimited sick days)
Annual equipment budget (once 2 month probation has been completed)

CompensationNegotiable

Required profile