Senior AI Engineer, Agent Runtime & LLM Systems

Datasnipper; New York (NY); Hybrid; USD 180,000 - 260,000; Full time

Benefits offered by this job

Hybrid work (3 days onsite NYC)

Stock options

Competitive salary

Medical and dental coverage

Job summary

DataSnipper, the Agentic Automation Platform for audit and finance, seeks a Senior Applied AI Engineer to join the Agent Runtime team behind Alwin. You will shape the base agent layer, model access, and tool orchestration, collaborating with product teams to deliver auditable workflows that scale to hundreds of thousands of professionals.

You will own long-running agent processes, fight hallucinations, latency spikes, and optimize for cost and performance while maintaining a high bar for

Qualifications

• 5+ years of software engineering experience.
• Experience shipping LLM-powered products in production.
• Hands-on experience building agentic systems and tool orchestration.
• Experience with evaluation: datasets, offline/online experiments.
• Fluency with LLM APIs across multiple providers.
• Strong fundamentals in software architecture and delivery quality.
• Excellent communication with auditors and product partners.

Responsibilities

• Own the base agents: system prompts, memory and state management for long-running tasks.
• Design how agents use the tool layer, including extraction, retrieval, sandboxes, and MCP integrations.
• Implement agentic patterns: sub-agent composition, planning modes, and model routing.
• Ship end-to-end: prototype to production on Alwin, with audit-domain validation.
• Define and build automatic processes to improve agents over time.
• Keep costs under control and optimize AI-centric workflows.
• Instrument agent behavior with observability and production traces.

Skills

Python

LLM APIs

Agentic systems

Evaluation

Software architecture

Communication

Job description

DataSnipper, the Agentic Automation Platform for audit and finance, seeks a Senior Applied AI Engineer to join the Agent Runtime team behind Alwin. You will shape the base agent layer, model access, and tool orchestration, collaborating with product teams to deliver auditable workflows that scale to hundreds of thousands of professionals.

You will own long-running agent processes, fight hallucinations, latency spikes, and optimize for cost and performance while maintaining a high bar for