Software Engineer, Agentic AI

Dotmatics • Boston, MA

17h • Remote

About The Position

Dotmatics is seeking an experienced Software Engineer (Agentic AI) to build and own the AI agent at the core of the Dotmatics Luma platform. The role covers the agent’s reasoning, tool use, memory, failure recovery, and evaluation. The engineer will work in a small, collaborative team alongside Product, scientists, and the wider Dotmatics organization.

This is an AI-native engineering role. Coding agents are the default implementation path, while the engineer remains accountable for architecture, correctness, security, and review. The role emphasizes decomposing work for coding agents, giving them suitable context and constraints, verifying their output, and shipping production software efficiently without compromising quality.

The engineer will develop the agent and its harness, tools, state persistence, security controls, and supporting infrastructure. They will evaluate and improve agent behavior on real multi-step tasks, and build services and interfaces around the agent using Python, Node.js/TypeScript, and React. The position is remote-friendly. Dotmatics describes its global team as more than 800 colleagues supporting customers in over 180 countries.

Requirements

• Around 10 years of software engineering experience; candidates with more or less experience are also welcome to apply.
• Hands-on experience building LLM-powered agents in production.
• Experience shipping at least one LLM-powered agent to real users that handles multi-step tool use, error recovery, and long-running tasks.
• Experience with context engineering, memory, and retrieval for a long-running LLM application, including explaining context-window contents and compaction and how coherence and retrieval quality were measured.
• Experience with agent security, including permissions, sandboxing, human approval gates, prompt-injection defenses, and audit logging tested against adversarial inputs.
• Experience building tools or MCP servers for model use and measuring tool selection and call accuracy.
• Production engineering experience with Python, type checking, automated tests, packaging in CI, Node.js/TypeScript, and React services and interfaces, including streaming output, monitoring, tracing, and managed configuration and secrets.
• Ability to use autonomous coding agents such as Cursor, Claude Code, Codex, or equivalent as the primary way of building software.
• Interest in life sciences, lab instrumentation, or automation is sought.

Nice To Haves

• A degree or higher education in Computer Science, Engineering, or a related technical field.
• Experience with agent frameworks such as LangGraph, or building custom harnesses from scratch.
• Working knowledge of Databricks and BI tools such as Sigma, Power BI, or Spotfire.
• Experience with observability for LLM systems, including tracing and cost and latency monitoring.
• Experience with fine-tuning, prompt optimization, or model evaluation.
• Authored skills, hooks, or MCP servers that made agent-driven delivery repeatable for a team.

Responsibilities

• Develop the agent loop, harness, and control flow connecting the model, tools, and state, including subagents, long-running tasks, checkpointing, and recovery.
• Engineer context for the agent, determining what enters its context window, when to compact or summarize, and how to maintain coherence during long-running experiments.
• Design agent-facing tools and interfaces for instrument orchestration, scheduling, data capture, and developer tooling such as GitHub, repositories, CI/CD, and workflows.
• Implement memory and state persistence so the agent can build on earlier work.
• Implement permissions and sandboxing, including boundaries for autonomous actions, human approval, and containment of mistakes.
• Defend the agent against prompt injection and untrusted content, with auditability and data isolation suitable for regulated environments.
• Evaluate the agent end to end on real multi-step tasks; analyze traces, find root causes, and improve the harness, tools, or prompts to improve reliability, tool-use accuracy, cost, and latency.
• Adapt the harness as models improve and remove scaffolding that is no longer needed.
• Run an AI-first development loop: scope, plan, delegate to coding agents, verify, and ship; prefer agent execution over hand-writing when quality holds.
• Build and maintain the team’s agent harness for development, including project memory, skills, hooks, MCP, and CI checks, enabling safe, bounded engineering work by agents.
• Measure and improve AI-driven development through throughput, defect rate, and review cost, and coach others on AI-native practices.
• Set up architecture and infrastructure for reliable agent operation, and manage configuration, environments, and secrets.
• Deliver services and interfaces around the agent in Python, Node.js/TypeScript, and React, including streaming output and extension points such as MCP.

Benefits

• Medical, dental, and vision insurance.
• Health Spending Accounts.
• Wellness benefits, including mental health apps and fitness perks.
• Company-paid life and disability insurance.
• 401k retirement plan with a 4% company match and immediate vesting.
• Flexible Time off Plan for US exempt employees.
• Certain positions are eligible for variable pay; details are discussed with the recruiter.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed