M
Applied AI Engineer – NYC Prop Trading Firm – Tier 1 Firm
Mondrian Alpha
2 hours ago
Full-time
On-site
East New York, New York, United States
About the Role
The Applied AI Engineer turns research and off-the-shelf models into systems the business actually runs on, pricing, risk, research workflows, or internal tooling. If a model ships but never gets integrated into a real workflow, it doesn't matter how good it is; this role is the difference between a promising prototype and something a portfolio manager touches every day.
What You'll Do Design, build, and ship production AI features end to end, from model selection through integration and monitoring Fine-tune and evaluate LLMs and other models against real business tasks, not generic benchmarks Build the evaluation harnesses and guardrails needed to trust a model's output in a production setting Work directly with researchers, PMs, and engineers to translate a vague idea into a shipped tool Own the full lifecycle of what you build: performance, cost, reliability, and iteration
Must-haves 3+ years shipping production software, with hands-on experience building applied AI or ML-backed features, not just research notebooks. This is non-negotiable for the role Strong Python and solid engineering fundamentals: testing, version control, code review Practical experience with LLM APIs, fine-tuning, or retrieval-augmented systems Comfortable owning a project from ambiguous spec to production
Nice-to-haves Experience in a regulated or high-stakes environment (finance, healthcare) where correctness and auditability matter Familiarity with vector databases, embeddings, or agentic workflows Prior exposure to a quant or trading environment Open-source contributions or public writing on applied AI work
Why This Role This is a seat where the work ships fast and gets used immediately, not a research role that ends at a slide deck. The team is small enough that one engineer's judgment on what to build, and what not to, actually shapes the roadmap.
What You'll Do Design, build, and ship production AI features end to end, from model selection through integration and monitoring Fine-tune and evaluate LLMs and other models against real business tasks, not generic benchmarks Build the evaluation harnesses and guardrails needed to trust a model's output in a production setting Work directly with researchers, PMs, and engineers to translate a vague idea into a shipped tool Own the full lifecycle of what you build: performance, cost, reliability, and iteration
Must-haves 3+ years shipping production software, with hands-on experience building applied AI or ML-backed features, not just research notebooks. This is non-negotiable for the role Strong Python and solid engineering fundamentals: testing, version control, code review Practical experience with LLM APIs, fine-tuning, or retrieval-augmented systems Comfortable owning a project from ambiguous spec to production
Nice-to-haves Experience in a regulated or high-stakes environment (finance, healthcare) where correctness and auditability matter Familiarity with vector databases, embeddings, or agentic workflows Prior exposure to a quant or trading environment Open-source contributions or public writing on applied AI work
Why This Role This is a seat where the work ships fast and gets used immediately, not a research role that ends at a slide deck. The team is small enough that one engineer's judgment on what to build, and what not to, actually shapes the roadmap.