Senior AI Engineer - Model Selection and Orchestration

zipcolimited; États-Unis; À distance; USD 162 000 - 205 000; Plein temps

Avantages offerts par ce poste

Unlimited PTO

Generous parental leave

Employer 401(k) match

Learning and wellness stipend

Health insurance for employees and dep

Remote-first policy with Manhattan in‑

Résumé du poste

zipcolimited is seeking a senior software engineer to build production-grade AI systems, focusing on the decision and orchestration layer that guides AI task execution. The role spans customer-facing AI experiences and internal agent workflows in a financially regulated setting, with remote-first policy for US-based employees and an option to work from a Manhattan office.

Responsibilities include designing routing logic, implementing intent classification, benchmarking approaches, and building

Qualifications

• 10+ years of professional software engineering experience.
• Proficiency in Python or TypeScript with API design and automated testing.
• Experience with LLM tool calling, structured outputs, retrieval-augmented generation, and agent/workflow orchestration.

Responsabilités

• Design the routing layer to decide whether requests are handled by deterministic logic, retrieval, a classifier, a smaller or reasoning model, a tool call, or human escalation.
• Build intent classification and clarification flows, evaluate confidence, and identify ambiguous input per task.
• Benchmark rules, retrieval approaches, and models using quality, latency, safety, and cost metrics to guide selection.
• Implement model adapters, structured outputs, timeouts, retry strategies, and fallback behavior in a testable orchestration layer.
• Connect agents to approved APIs and tools with authentication, authorization, policy checks, and idempotency.
• Partner with security to implement input/output protections and prompt-injection defenses.
• Develop reusable orchestration patterns and components to aid other teams.
• Measure and improve AI behavior in production through structured evaluation.

Connaissances

Python

TypeScript

APIs

Automated testing

Distributed services

CI/CD

LLM tool calling

Agent orchestration

Security & IAM

Description du poste

Role overview

This is a senior engineering role focused on building production-grade AI systems, with particular emphasis on the decision and orchestration layer that determines how AI tasks should be executed. The work spans customer-facing AI experiences and internal agent workflows in a financial technology setting, where reliability, security, and sound engineering judgment are fundamental. The position is remote-first for US-based employees, with the option to work in-person from a Manhattan office.

Responsibilities

• Design and implement the routing layer that decides whether a request should be handled by deterministic logic, retrieval, a classifier, a smaller or reasoning model, a tool call, or human escalation.
• Build intent classification and clarification flows that distinguish information requests from actions, evaluate confidence, and identify ambiguous input.
• Benchmark rules, retrieval approaches, and models against representative tasks, using quality, latency, safety, and cost metrics to guide selection.
• Implement model adapters, structured outputs, timeouts, retry strategies, and fallback behavior in a testable orchestration layer that avoids over-reliance on any single model provider.
• Connect agents to approved APIs and tools with appropriate authentication, authorization, policy checks, and customer confirmation, handling duplicate requests, idempotency, and recovery for state-changing actions.
• Partner with security and risk functions to implement input and output protections, sensitive-data handling, prompt-injection defenses, and controlled access.
• Develop reusable orchestration patterns, adapters, and routing components that make it easier for other engineering teams to build AI-powered experiences.
• Measure and improve AI behavior in production through structured evaluation work.

Requirements

• 10+ years of professional experience building, testing, deploying, and operating production software, including hands-on delivery of AI or ML-enabled applications beyond prototypes.
• Strong proficiency in Python or TypeScript, with modern software engineering practices including API design, automated testing, distributed services, and debugging across application, model, and tool boundaries.
• Practical experience building applications using LLM tool calling, structured outputs, retrieval-augmented generation, and agent or workflow orchestration.
• Ability to evaluate classification quality, uncertainty, fallback strategies, and different model or system approaches using measurable outcomes such as task success, latency, reliability, and total inference cost.
• Experience with cloud deployment, automated testing, CI/CD, observability, identity and access controls, secrets management, retries, idempotency, and failure recovery in production systems.
• Strong judgment about when a problem should be solved with deterministic software, retrieval, a model, or a combination, and when an agent should ask for clarification or hand work to a person.

Benefits and work setup

• Remote-first role for US-based employees, with the option to work in-person from a Manhattan office.
• Flexible working culture and incentive programs.
• Unlimited paid time off, generous paid parental leave, and family support policies.
• Company-sponsored retirement plan with employer match.
• Learning and wellness subscription stipend.
• Employer-sponsored insurance for employees and dependents, with several fully covered options.
• Reported annual base pay range of $162,000 to $205,000, with additional premium percentages possible based on a tiered premium strategy; bonus and equity may also be part of the total compensation package.