Agentic AI Engineer
Obin AI
- Location
- Remote (NYC-Metro, New York)
- Compensation
- $150k - $180k/yr
- Employment
- Full-time
- Level
- Senior Level
Posted Today
About the role
Obin AI is building an agentic workforce for regulated financial institutions, with agents already managing over $1 trillion in assets. This role involves owning technical decisions to build and optimize multi-agent systems that automate complex credit workflows, balancing accuracy, latency, and cost-efficiency.
Skills
Python
LLMs
RAG
Vector Databases
GCP
APIs
Multi-agent orchestration
Agent-to-agent handoffs
Vector DB routing
Inference cost optimization
Eval frameworks
Guardrail frameworks
Human-in-the-loop systems
Technical architecture
Infrastructure deployment
Perks
- Equity
- Hybrid Work
Full job details
Have you taken a multi-agent system into production and then had to make it faster and cheaper? Want to do that where every output has to be explainable to a credit firm?
This isn't your typical senior engineering seat. Obin AI builds AI workers for regulated finance, and our agents already run inside institutions managing over $1 trillion in assets.
You will work directly with the founders and with customers. You will own key technical decisions from day one.
Beyond what brings you to Obin, this is what your day to day could look like.
- Title: Senior AI Engineer
- Location: New York, hybrid
- Core Tech Stack: Python, LLMs, RAG, vector databases, GCP (preferred), APIs
- The Mission: Build agentic systems that automate complex credit workflows. Balance accuracy, latency, reliability, cost and explainability, and lead the experimentation cycles that improve them.
- Platform scope: Define technical architecture and build reusable infrastructure for agent deployment.
- Customer scope: Partner with stakeholders at major credit firms to find workflow pain points and co-develop solutions tied to ROI.
- Compensation: $150,000 to $180,000 base + equity
What we need: 6 to 8+ years of software engineering and 1 to 2+ years building LLM-based agents or RAG systems in production. Experience with multi-agent orchestration, agent-to-agent handoffs, vector DB routing, and optimizing LLM latency and inference cost. You have contributed to eval or guardrail frameworks.
Experience with human-in-the-loop systems or long-horizon agentic tasks is a plus.
Ready to change the future of Finance? Apply today for an insight into how...