The Role

Build the AI layer underneath the science

Terramtech is looking for an Applied AI Engineer — Agentic Systems to build the agentic systems that run our science, manufacturing, regulatory, and commercial operations.

You will design, build, deploy, and continuously improve production-grade AI agents that automate complex scientific, operational, regulatory, manufacturing, and commercial workflows.

Working directly alongside Terramtech’s Founding AI Agent Engineer, you’ll help establish the technical foundation for the company’s AI platform — partnering closely with scientists, engineers, operations leaders, and executives to translate manual, judgment-intensive work into intelligent agentic systems.

The Basics

At a glance

• DepartmentEngineering / Artificial Intelligence
• Reports toFounding AI Agent Engineer
• LocationHybrid — Phoenix, Arizona
• Employment typeFull-Time
• CompensationCompetitive base salary + meaningful early-stage equity
• ScopeCompany-wide — research through commercial

About Terramtech

Better science, not better marketing

Terramtech is redefining sustainable packaging by replacing traditional plastic barriers with bio-based materials that integrate seamlessly into existing paper packaging manufacturing processes. Our technology delivers the barrier performance, food safety, manufacturability, and commercial scalability required for global adoption — while significantly reducing environmental impact.

As we scale from pilot manufacturing to commercial production, we’re building intelligent systems that let our scientists, engineers, and business teams operate faster and more accurately. AI is becoming a core capability across every function of the company.

What You’ll Do

What you’ll build

AI platform development

• Design and develop autonomous and semi-autonomous AI agents supporting research, manufacturing, regulatory, quality, commercial, and corporate operations.
• Architect multi-agent workflows using planning, memory, retrieval, orchestration, and tool use appropriate to each business process.
• Develop production-quality software integrating large language models with internal applications, APIs, laboratory systems, ERP platforms, and scientific databases.
• Build reusable AI services and infrastructure supporting scalable enterprise deployment.

Scientific knowledge systems

Partner directly with formulation scientists and process engineers to:

• Capture scientific reasoning and decision-making processes.
• Translate laboratory workflows into agent behavior.
• Build retrieval systems over experimental data, technical literature, lab notebooks, and regulatory documentation.
• Improve access to institutional knowledge through AI-powered search and reasoning.

Manufacturing & operations automation

• Monitor manufacturing process data and detect anomalies and quality drift.
• Surface root-cause hypotheses and support production decision making.
• Reduce manual data movement across business systems.

Regulatory & compliance automation

• Draft regulatory documentation and compile FDA and international food-contact submissions.
• Identify documentation gaps and organize technical records.
• Improve regulatory readiness and audit support.

Commercial enablement

• Match customer packaging specifications to Terramtech formulations.
• Generate technical summaries and assist sample requests.
• Accelerate customer onboarding and improve sales engineering efficiency.

AI reliability & governance

• Build evaluation frameworks to measure AI performance.
• Implement guardrails and human-in-the-loop review.
• Improve hallucination resistance through retrieval and structured data.
• Ensure transparency, explainability, and auditability.
• Continuously optimize model quality, latency, and operating cost.

Production operations

• Own AI systems after deployment: production health, observability, cost, latency, failure triage, and documentation.

Platform strategy

• Partner with the Founding AI Agent Engineer to establish engineering standards, evaluate emerging AI technologies, make build-vs-buy calls, define long-term architecture, and help scale the AI Engineering organization.

What We’re Looking For

Minimum qualifications

• Bachelor’s degree in Computer Science, Software Engineering, AI, Machine Learning — or an equivalent combination of education and experience.
• 5+ years of professional software engineering experience building production software.
• Expert-level proficiency in one or more modern general-purpose languages, with the ability to work fluently across languages. (We work primarily in Python and TypeScript.)
• Experience building production systems with REST APIs, asynchronous/event-driven processing, relational databases, queues, and cloud infrastructure. (We run on AWS.)
• Demonstrated experience building and shipping production AI agents / LLM-powered applications.
• Hands-on experience across the modern agent toolchain — agent frameworks, managed LLM inference, tool use / function calling, and multi-agent orchestration (e.g. LangGraph, AWS Bedrock, Model Context Protocol, or equivalents). We build primarily on Anthropic Claude and value experience with any modern LLM provider.
• Experience designing Retrieval-Augmented Generation (RAG) systems: embeddings, chunking strategies, retrieval optimization, and context engineering.
• Strong prompt engineering and evaluation methodology experience — labeled eval sets, offline eval and regression gating.
• Experience productionizing LLM systems responsibly: guardrails, human-in-the-loop oversight, fallback logic, observability, and testing, with attention to reproducibility and determinism.
• Strong software engineering fundamentals: version control, testing, debugging, documentation, and maintainable architecture.
• Excellent communication skills, with the ability to collaborate across technical and non-technical teams.

Preferred qualifications

Experience in one or more of the following is highly desirable:

• Manufacturing, supply chain, or ERP integration
• Packaging
• Enterprise AI platform development
• Compliance-heavy or regulated environments (SOC 2 / SOX, IPO-track, FDA, food safety)
• MLOps / LLMOps and model evaluation frameworks
• Startup or high-growth technology companies
• Materials science, industrial automation, or scientific computing
• Fine-tuning
• Lab-adjacent systems: LIMS, scientific instrumentation, experimental data analysis, chemical engineering

Why Join

AI here is not a side project

At Terramtech, artificial intelligence is becoming a foundational capability of how we innovate, manufacture, and scale sustainable materials. Our AI platform already supports real production work across the company.

You’ll help shape the architecture, standards, and future of AI at Terramtech — working directly with leadership and subject matter experts on meaningful scientific and operational challenges.

If you’re excited about building trustworthy AI systems that have measurable impact on sustainability, manufacturing, and material science, we’d love to meet you.

Apply

Show us your work

No algorithm puzzles, no gotcha questions. Send us whatever is most relevant — something you’ve built, written, run, or solved. Four steps: intro call, show us your work, a real scoped exercise, final conversation.