Senior Applied AI Engineer – Agent Runtime (Alwin by DataSnipper)
About this role
The team at DataSnipper is developing Alwin, an Agentic Automation Platform designed for audit and finance. This platform creates agents capable of executing complex, multi-step audit procedures by tracing every output back to its original source evidence, requiring final human review and sign-off. The core component this role supports is the Agent Runtime, which serves as the foundational layer for all agents. This runtime encompasses model access, system prompts, tool integration, agent orchestration, and observability mechanisms that validate agent performance against defined objectives.
This Senior Applied AI Engineer will focus on the applied AI aspects of this runtime. The individual will determine how base agents reason, which tools they utilize, the level of abstraction for context management over extended periods, and how accuracy, speed, and cost are measured. The work is situated in a small team where the impact of the work is significant, affecting hundreds of thousands of audit and finance professionals. A successful candidate will contribute to establishing production standards for agents in a regulated domain where existing playbooks are scarce.
This role is ideal for someone who thrives on solving open-ended production problems. It requires a blend of deep engineering skills and a strong ability to communicate technical decisions to domain experts. Candidates who enjoy defining the structure of AI systems and improving reliability under real-world constraints would find this environment suitable.
What you will do
- Own the core components of the base agents, including system prompts, context engineering, and memory management for tasks spanning many execution turns.
- Design the mechanism for agents to interact with the tool layer, deciding on the appropriate level of abstraction for document extraction, retrieval, and sandbox execution.
- Implement advanced agentic patterns such as sub-agent composition, planning modes, human-in-the-loop gates, and model routing across different providers.
- Manage the entire lifecycle of agent development, taking work from initial prototype with domain experts through evaluation to final production deployment on Alwin.
- Develop and maintain automated processes aimed at continuously improving the agent's performance over time.
- Analyze long agent trajectories to extract meaningful signals, attribute outcomes to specific reasoning steps, and convert failure modes into actionable fixes.
- Establish metrics for accuracy, latency, and token cost, clearly articulating trade-offs to downstream product teams.
What you need
- Possesses at least five years of software engineering experience, backed by strong, production-grade Python skills.
- Demonstrates experience shipping and operating LLM-powered products in production, including managing issues like hallucinations, latency spikes, and cost management at scale.
- Has hands-on experience building agentic systems, focusing on control loops, tool selection, planning versus execution logic, and handling retries and fallbacks beyond simple prompt-and-parse pipelines.
- Requires experience in evaluation methodologies, including building datasets, running both offline and online experiments, and using results to measurably improve AI systems.
- Must be fluent with LLM APIs and agent frameworks across multiple model providers.
- Should possess strong fundamentals in software architecture and system design, coupled with a proven track record of delivering reliable and well-tested software.
- Requires excellent communication skills necessary for collaborating directly with auditors and product partners to define the criteria for correctness.
Nice to have
- Experience with Durable workflow engines or long-running background task systems
- Experience with RAG and retrieval pipelines over large, messy document corpora
- Document AI: VLMs, OCR, structured extraction and their metrics
- Sandboxed code execution, MCP, or multi-agent orchestration in production
- Domain experience in audit, accounting or fintech
- Familiarity with OWASP GenAI security practices and working in a regulated, privacy-sensitive environment
Benefits
- $230K – $300K salary
Skills and tech stack
Questions candidates ask
What is the remote policy?
The posting does not explicitly state the remote work policy for this role.
What is the compensation range for this position?
The salary offered for this role is between $230K and $300K.
What is the primary technology stack used?
The core technologies mentioned include Python, LLM APIs, and various agent frameworks.
How is collaboration structured within the team?
The role requires partnering with Agent Experience teams, Document Intelligence, and the AI Platform team to translate needs into self-service runtime capabilities.