Description:

We are building Reflow, a workforce and workflow intelligence platform that helps teams understand and improve how work gets done.
A key part of that vision is Reflow AI, which focuses on agentic, automation-driven intelligence that adapts to how people actually work.
We are looking for an AI developer who can experiment quickly, optimize intelligently, and help us push the boundaries of AI-assisted performance.
The responsibilities include developing, optimizing, and maintaining agentic AI pipelines for automation and workflow intelligence.
You will experiment with prompt engineering, context tuning, and model orchestration to improve performance and adaptability.
The role involves working with vector databases, RAG (retrieval-augmented generation), and related frameworks for contextual intelligence.
You will collaborate with engineers and product teams to integrate AI systems into real-world workflows.
Continuous refinement and improvement of models through experimentation, data feedback, and iterative development is expected.

Requirements:

A strong foundation in Python, LLMs, and agentic architectures is required.
Experience in prompt engineering, context optimization, and AI pipeline design is necessary.
You should be comfortable building fast prototypes and testing model behavior in varied, complex contexts.
A curious, pragmatic attitude and the ability to experiment are essential, as you will need to turn uncertainty into working systems.

You will help shape the intelligence core of Reflow, learning from work patterns and turning data into action.
The role offers the opportunity to work across experiments, prototypes, and production systems, collaborating directly with founders and engineers.
The position is flexible regarding structure, allowing for part or full-time work.
What matters most is iteration speed, creativity, and a bias for working code.

Skills