Applied AI Engineer
Astraeus is building AI-native infrastructure for wealth and financial services. We are not adding a thin AI layer to legacy software. We are building systems where models, tools, data, and workflows come together to produce real analytical and operational leverage in production. Our focus is on agentic analytics, intelligent workflows, and execution systems that can operate inside complex, high-stakes environments with real constraints around security, compliance, auditability, and reliability. We are looking for an exceptional Applied AI Engineer to help build these systems from the ground up.
The Role
You will design and ship production-grade AI systems that turn modern model capabilities into useful, durable products. This includes agentic analytics, tool-using workflows, retrieval and context systems, orchestration, evaluation, and the supporting infrastructure needed to make these systems reliable in the real world.
This is a high-ownership role for someone who wants to work close to the frontier, but cares as much about correctness, observability, and security as they do about raw capability. You should be excited by fast-moving applied AI, but grounded enough to know that production systems win on discipline, not demos.
You will be expected to move quickly, own major surface area, and help define both architecture and product behavior. This is not a narrow implementation role. It is a foundational engineering role for someone who wants to build durable AI systems in production in a highly regulated industry.
What You’ll Do
- Build production AI systems for analytics, research, workflow automation, and decision support
- Design and implement agentic workflows that use models, tools, retrieval, memory, and structured data together
- Develop robust orchestration patterns for multi-step reasoning and action-taking systems
- Improve system quality through evals, tracing, failure analysis, guardrails, and feedback loops
- Build for reliability, observability, and auditability from the start
- Work across the stack, from backend systems and APIs to model integration and workflow design
- Collaborate closely with leadership and product on what to build, not just how to build it
- Stay current on the latest applied AI and agentic engineering techniques, and translate useful advances into production improvements
- Help establish engineering standards around security, testing, reviewability, and operational safety
What We’re Looking For
- Strong engineering fundamentals and a track record of shipping real software
- Experience building with modern LLM systems, agent workflows, or AI-powered product features
- Strong understanding of the practical challenges in applied AI: context management, tool use, evals, failure modes, and system reliability
- High agency and comfort operating in ambiguous, fast-moving environments
- Strong product instinct and willingness to own outcomes end-to-end
- Security-minded engineering judgment and respect for working with sensitive data and constrained environments
- Ability to balance speed with rigor
- High standards, low ego, and strong intellectual honesty
- Experience with agentic systems, orchestration frameworks, retrieval pipelines, or custom AI infrastructure
- Experience building analytics or decision-support products
- Experience in fintech, enterprise software, or other data- and workflow-heavy environments
- Experience with evaluation systems, observability, and testing for LLM-based applications
- Familiarity with designing systems for human review, approval, and intervention
- Comfort working across backend, infra, and applied AI rather than staying inside a narrow specialization
Strong signals
- You have built AI systems that moved beyond demos into production use
- You know how to separate raw model capability from production readiness
- You are comfortable working across models, tooling, backend systems, and workflow design
- You care about evals, failure modes, observability, and guardrails as much as raw output quality
- You can move quickly without becoming sloppy
- You are serious about craft, but low ego in how you work with others
You might be a fit if you
- Want to build consequential AI systems, not just experiment around the edges
- Like hard problems, short feedback loops, and real accountability
- Naturally take responsibility beyond your formal scope
- Can go from prototype to production without losing the thread
- Are excited by fast-moving technical change, but can separate durable ideas from hype
- Want meaningful room to shape architecture, product behavior, and engineering standards early
Why Astraeus
- You’ll help define the architecture, product, and engineering culture of an AI-native company at an early stage
- You’ll work on real problems in high-stakes environments with real constraints
- You’ll have unusual ownership and responsibility from day one
- You’ll build systems that are meant to matter in production, not just in demos
- You’ll work with a small, high-trust team with high standards and low politics
Introduce Yourself
If your expertise aligns with Astraeus but doesn't map to a current open role, we want to hear from you.
We're particularly interested in domain experts in wealth management operations, compliance architecture, data engineering, agentic systems, and enterprise sales to RIAs and private equity firms.
Tell us what you've built, what you understand about the problem we're solving, and how you'd contribute. We review every submission personally.
We're particularly interested in domain experts in wealth management operations, compliance architecture, data engineering, agentic systems, and enterprise sales to RIAs and private equity firms.
Tell us what you've built, what you understand about the problem we're solving, and how you'd contribute. We review every submission personally.