AI Engineer – Generative AI, Agents

This is a fully remote position, open to applicants in Argentina, +5 more countries.

📋 Description

• Design and implement retrieval systems utilizing chunking and embedding pipelines, hybrid search methodologies, reranking, and retrieval-quality assessment.

• Develop advanced, multi-step agentic workflows that incorporate tool calls, MCP servers, structured output mandates, context-window management, deterministic fallbacks, and human-in-the-loop checkpoints.

• Construct evaluation test sets, execute model-as-judge scoring, track regressions, and perform error analyses.

• Establish defenses against prompt injection, conduct output validation, implement guardrails, handle PII, and ensure graceful degradation.

• Deploy containerized applications on Azure or AWS, utilizing CI/CD practices and ensuring observability.

• Manage budgets for latency, cost, tokens, and throughput.

• Collaborate within client repositories and environments, participating in standups and occasional customer interactions.

• Create production-ready AI systems that are dependable, quantifiable, and cost-effective for end-users.

⛳️ Requirements

• Over 4 years of experience in developing and delivering production software, with a modern backend language (such as Python) as your primary focus.

• Proficient in testing, code reviews, CI/CD practices, Git, containerization, API design, and asynchronous programming.

• Practical experience with LLM-based systems, including retrieval-augmented generation, function and tool calling, structured output enforcement, and prompt engineering.

• Hands-on knowledge of tools such as pgvector, Pinecone, Qdrant, FAISS, or Azure AI Search.

• Familiarity with LangGraph, LangChain, CrewAI, MCP, or native Python execution loops.

• Developed an evaluation suite for an LLM system, inclusive of test-set design, model-as-judge or equivalent scoring, and regression tracking.

• Experience in cloud deployments using Azure or AWS, Docker, CI/CD pipelines, and infrastructure-as-code tools like GitHub Actions, Terraform, or Bicep.

• Ability to articulate trade-offs regarding latency, token costs, and throughput.

• Regular use of Claude Code, Cursor, or GitHub Copilot in actual delivery projects.

• Proficient in written and spoken English, at C1 level or higher.

• Bachelor's degree in Computer Science, Data Science, or a related field, or equivalent professional experience.

• Preferred: experience in fine-tuning with LoRA, QLoRA, and PEFT; self-hosted/open-weight inference; multimodal systems; LLM security; SOC 2 or HIPAA compliance; streaming or high-throughput inference; contributions to open-source AI or technical writing.

🏝️ Benefits

• Fully remote-first culture (work from anywhere in Latin America).

• Paid time off (PTO).

• U.S. Holidays.

• AI training and certification opportunities.

• Mentored career development.

• Profit sharing.

• Compensation in U.S. dollars.