Sobre esta vaga de AI Engineer (Brilliant Harvest) na Altaml
About the Role
As an AI Engineer, you will build and ship production-ready AI agents and features for the Brilliant Harvest platform. Working closely with the Lead AI Engineer and the broader AI team, you'll own well-scoped features end to end. Your work will center on agents that plan, make decisions, and take action, backed by the RAG pipelines, document ingestion, and models that power them.
This role is ideal for an engineer with a few years of experience who thrives in a fast-paced environment and loves solving complex problems. You'll take ideas from concept to deployment in a domain where AI is transforming a legacy industry. Along the way, you'll get hands-on mentorship and room to grow into larger, more ambiguous problems.
Key Responsibilities
Agents
Design, build, and ship production AI agents that plan, use tools, and complete multi-step tasks across the Brilliant Harvest platform.
Implement agent orchestration, memory, and tool integrations (e.g., MCP servers), with guardrails and human-in-the-loop checkpoints for high-stakes actions.
Evaluate agents end to end, measuring task success, tool-call accuracy, cost, and latency, not just final answers.
Build & Ship
Design, develop, test, document, and maintain production features for our AI platform, including the RAG pipelines and LLM integrations our agents rely on.
Build and improve components of our document ingestion pipelines (parsing, chunking, enrichment, and indexing), ensuring accuracy and compliance with manufacturer standards.
Investigate agent and answer-quality issues, tracing them back through tool calls, retrieval, and ingestion to find the root cause.
Help build and maintain evaluation datasets and automated checks, including LLM-as-judge, so regressions are caught before customers see them.
Prototype with new LLMs, agent frameworks, and tooling, and share findings that inform model and vendor decisions.
Engineering Practices
Participate in brainstorming sessions, design reviews, and code reviews, giving and receiving thoughtful feedback.
Write clean, well-tested code, follow the team's coding standards and CI/CD practices, and suggest improvements where you see gaps.
Use AI-first development workflows (e.g., Claude Code or similar agentic tools) to work faster, while critically reviewing everything the AI generates.
Help analyze requirements for your features, break them into well-defined tasks, and provide reliable effort estimates for your own work.
Cross-Functional Collaboration
Work collaboratively with Product, Data, Design, and QA teams to deliver features aligned with business objectives and the product roadmap.
Provide input on the technical feasibility of UI/UX designs for AI features, and help integrate them with backend systems.
Communicate progress, blockers, and technical tradeoffs clearly to both technical and non-technical teammates.
Learning & Knowledge Sharing
Grow through hands-on mentorship from the Lead AI Engineer, taking on larger and more ambiguous problems as you're ready.
Share what you learn through code reviews, pairing, documentation, and demos, and help junior engineers get up to speed.
Embrace a "Systems Thinking" mindset, moving from writing every line by hand to orchestrating and critically reviewing AI-generated solutions.
Qualifications & Skills
3+ years of professional software engineering or data science experience, including at least 1 year building AI/ML or LLM-powered applications.
Practical experience building LLM agents, including tool calling, multi-step orchestration, and agent frameworks or protocols such as MCP.
Hands-on experience with generative AI and LLMs, such as RAG, embeddings and vector search, or document ingestion, ideally with at least one feature in production.
Strong proficiency in Python, plus working experience with at least one of: TypeScript, C# (.NET), or another modern backend language.
Working knowledge of API design, microservices, and cloud infrastructure.
A collaborative approach: you communicate clearly, raise blockers early, and give and receive feedback constructively.
Familiarity with React, React Native, or other modern frontend frameworks, and how they integrate with AI-powered backends, is an asset.
Experience with prompt engineering best practices and AI-first development tooling (e.g., Claude Code or similar) is a strong asset.
Exposure to evaluation methods for LLM applications and agents, such as test sets, LLM-as-judge, or human review, is an asset.
Experience working in a startup environment is an asset.
Familiarity with the agriculture, heavy equipment, or dealership domain is a strong asset.
What Will Set You Apart?
You've built an agent that does real work, not just a demo. You know agents usually fail because of vague tools, missing context, or no way to recover, and you design for that.
You're comfortable when the roadmap moves under your feet. You turn loosely defined problems into working software quickly, and you're building the judgment to know which shortcuts will cost you later.
You've shipped something real with LLMs, at work or in a serious side project. You've seen what changes once real users, messy data, and latency or cost limits show up.
You know RAG quality starts with retrieval. When an answer is wrong, your first move is to check whether the right document surfaced, not to rewrite the prompt.
You treat evals as part of the job, not an afterthought. You've written test cases or LLM-as-judge checks, and you'd rather catch a regression in CI than from a customer.
You've experimented with decision models like TypeSafe's Jev, which return typed choices and scores with probabilities instead of text. You know when a fast, cheap decision call beats an LLM, such as classifying a document or choosing an agent's next step.
Why Join Us?
Be part of a high-performing team led by Remi Schmaltz, an entrepreneur with decades of experience launching and growing agriculture businesses.
Remote-first role with a flexible work environment.
A front-row seat to how AI is changing the way equipment dealers, farmers, and contractors work.
A collaborative culture that values growth, learning, and impact.
Competitive compensation, ESOP, and benefits.
