Work on problems
that haven't been solved.
Syntarus builds production memory infrastructure for AI agents. Our work spans retrieval, knowledge graphs, data governance, developer tooling, multi-agent coordination, and the operational systems required to run them reliably.
What it's like here
We're a small, high-conviction team. We move quickly, think carefully, and hold our work to a high bar - not because anyone is watching, but because we care about what we're building. You'll have direct ownership over research areas and production systems from day one. There's no bureaucracy between your ideas and their deployment.
We work at the intersection of research and production. That means you'll spend time on papers and also on debugging production failures at 11pm. Both matter. We don't separate "the people who think" from "the people who build." Everyone does both.
What we look for
Deep technical curiosity
You're drawn to hard problems for their own sake - not because they look good on a CV. You read papers that are adjacent to your work because you want to, not because someone told you to.
Bias for deployment
A solution that doesn't run in production doesn't count. You're comfortable in research mode and engineering mode - and you can switch between them in the same afternoon.
Comfort with ambiguity
We work on open research questions. The requirements are often "figure out why this doesn't work." You need to be able to make progress without a spec.
High standards, low ego
We push back on each other's work constantly and constructively. You need to be able to have your code and ideas challenged without taking it personally - and to challenge others the same way.
Open research & engineering roles
We're currently building our core team. If any of the following match where you want to focus, we want to hear from you - even if the description isn't an exact match for your background.
Agentic Memory Systems
Work on memory consolidation, retrieval, and contradiction resolution in long-horizon agent deployments. You'll be working directly on the systems that power production agents. Strong background in ML systems, vector databases, or knowledge representation preferred.
Multi-Agent Coordination
Design and study coordination protocols for multi-agent systems - planning, tool-use grounding, and failure recovery. You'll build systems that run in production and study why they fail. Background in distributed systems, RL, or LLM agent systems preferred.
AI Product Engineering
Build the products and infrastructure that put our research in front of users and enterprise clients. You'll own the full stack - from the LLM agent loop to the user interface - and be directly involved in shaping what we build and how it works.
Don't see your exact role?
If you're a strong researcher or engineer who cares about agentic AI and thinks you'd contribute meaningfully - write to us. The best applications we've received have been people who didn't fit a template.
Send a note to [email protected] with what you work on and what you'd want to work on here. No template required - we read everything.
Internship openings
We have current openings for the following internship roles. Interns at Syntarus work on real research problems and real production systems - not slide decks.
- AI/ML Research Intern - work on agentic memory or coordination research alongside our research engineers
- Full Stack Developer Intern - build features and infrastructure for our enterprise agent platform
- Marketing & Communications Intern - help us communicate technical research and products to the right audiences