Senior Software Engineer, Agents

Harvey; San Francisco (CA); On-site; USD 100,000 - 150,000; Full time

Job summary

Deepstreamtech is looking for a Software Engineer, Agents to build systems that enhance AI agents' effectiveness for legal professionals. You will design environments, make model selection decisions, and develop tools for faster processes.

The ideal candidate has 5-8 years of software engineering experience and is proficient in Python, with a strong background in shipping user-facing products. This role is crucial for improving legal workflows and enhancing agent performance.

Qualifications

• 5-8 years of software engineering experience.
• Experienced in shipping impactful products.
• Familiar with practical evaluations for task quality.

Responsibilities

• Build systems that make AI agents indispensable to legal professionals.
• Design environments and actions for agentic work.
• Optimize agent performance through various engineering practices.

Skills

Building effective domain-specific agents

Proficiency in Python

Working with LLM APIs

Experience with user-facing products

Job description

Requirements

• Passion for building effective domain-specific agents
• Iterative mindset: you develop proof of concepts, make decisions quickly, and ship v0s
• Comfortable with when and how to use evaluations to drive quality
• Humble and adaptable about code and frameworks. We expect you to drive adoption of new best practices as they develop
• 5-8 years (post-BS/MS) of software engineering experience
• We are hiring for this role across Mid, Senior, and Staff levels
• Proficiency in Python and experience working with LLM APIs and agent frameworks
• Experience with shipping user-facing products, either on the backend or full-stack

What the job involves

• As a Software Engineer, Agents, you'll build the systems that make our AI agents indispensable to legal professionals
• You will design environments and actions for agentic professional work, make model selection decisions, manage context windows, create optimal tools, and develop evals that enable faster iteration loops to unlock new capabilities
• We're looking for engineers and researchers who are immersed in the space, driven to ship impactful products, and are experienced in using practical evaluations to drive task completion quality and customer delight
• Save weeks of effort on high pressure M&A deals by automating information requests and diligence checks across hundreds of thousands of files with retrieval and file editing agents
• Dramatically speed up deal turnaround speed by improving the latency and quality of agents on applying a standard legal "playbook" to contracts
• Ensure that our clients win their complex, data-hungry litigation cases by optimizing our multi-source retrieval agents
• Enable board-ready PowerPoint generation by tuning the harness and libraries for coding agents
• Partner with customers and PMs to understand legal workflows, design practical evaluations that capture what "excellent" means, and ship agents that get the job done
• Optimize agent performance through prompt engineering, model selection, tool design, skill writing, context window management, and eval harness development
• Work with our model infra team to design and implement infrastructure for low-latency agent execution, including caching strategies, parallel tool calls, or subagent patterns
• Improve our observability and instrumentation to profile agent behavior, identify bottlenecks, and drive optimization decisions
• Stay current on new developments in agentic systems and bring those learnings back to the products we build