Full-Stack Engineer (AI-Enabled)

Role: Full-Stack Engineer (AI-Enabled)

Location: Mexico, Argentina – Remote

Employment Type: Contract

Responsibilities

• Build and run production AI agents by developing TypeScript agents and services on AWS (Lambda, Step Functions, DynamoDB, Bedrock) using Strands or LangGraph and AI SDKs.
• Design agent behavior by deciding what the model investigates and what plain code enforces.
• Sandbox and constrain agents by building read-only runtimes, scoped tool access, worktree-contained reads, and budget and timeout guards that remain effective when the model misbehaves.
• Write evaluations for AI-driven features by building golden sets and replays from real production cases, calibrating judges, and gating changes on recall, precision, and false-positive rates. Run model and effort-level comparisons before release.
• Make agents observable by emitting reason-coded metrics and structured logs, writing alarms for real failures, and investigating production misses in Splunk and VictoriaMetrics.
• Run infrastructure and pipelines by maintaining Terraform roots across environments under tagging and permission-boundary rules, and improving GitLab CI/CD for build, test, deploy, and quality gates.
• Handle integrations that withstand retries, rate limits, permanent delivery failures, and stale state.
• Use AI to write and review most code and keep agent context current.
• Rigorously review AI-written changes for security, correctness, and maintainability, catching issues that tests do not.
• Enable other teams by setting up assistant configuration, skills, hooks, and MCP servers; pairing with engineers; and turning successful approaches into reusable practices.
• Build the interfaces agents need through React and API development for review summaries, dashboards, and internal tooling, preferring outputs that agents, CLIs, and Slack can read.
• Document system behavior through sequence and architecture diagrams, runbooks, and short decision records explaining why.

Requirements

• LLM Features in Production: Experience shipping features using LLM APIs, including prompts, tool calling, structured output, and long context. Ability to describe the evaluation that gated a release, what failed, and what changed.
• TypeScript: Strong production experience with TypeScript and Node.js, including asynchronous code, testing, and monorepo tooling.
• Python: Working proficiency for evaluations, analysis, and scripting.
• AWS Serverless: Experience with Lambda, Step Functions, DynamoDB, IAM, event-driven design, and Bedrock.
• Infrastructure as Code: Ability to read and safely modify Terraform across environments.
• Data: Hands-on experience with NoSQL and SQL stores, including schema changes and migrations on live systems.
• Observability: Ability to design metrics, logs, and alarms for non-deterministic systems and debug issues using them.
• CI/CD: Proficiency with CI/CD tools, ideally GitLab CI/CD.
• AI-Assisted Development: Daily use of an AI coding assistant, with the discipline to verify its output.
• Communication: Fluent written and spoken English, with the ability to explain trade-offs and get a team to adopt an approach.

Nice to Have

• Multi-agent systems, MCP servers, or evaluation frameworks (e.g., Strands Evals).
• LLM cost control, including model routing, prompt caching, and token budgets.
• Prompt-injection defense and agent security.
• React and GraphQL for internal tools.
• Java and Spring Boot for reading or integrating with existing services.
• Code intelligence or retrieval work, including indexing, symbol search, and knowledge bundles.
• Kubernetes, AWS AppSync, and EventBridge.
• Playwright/Cucumber test automation.
• Claude Certified Developer or Claude Certified Architect (Anthropic).