Get a reply from this employer
Your resume and cover letter, tailored to exactly what they’re hiring for.
American AI Logistics is hiring a Software Development Engineer in Test (SDET) to own quality across a cloud-based B2B SaaS platform for government procurement workflows. You will design test strategies, build frameworks in TS/JS, and implement API, integration, and E2E tests across multiple services.
You’ll ensure data integrity, security, and scalability with CI/CD gates and performance testing. The role emphasizes testing AI-powered extraction and generation pipelines, non-deterministic
SDET, Quality Engineering & Test Automation
We’re looking for an SDET to build and own the quality engineering practice for a cloud-based B2B SaaS platform that automates U.S. government procurement workflows using AI. The platform ingests solicitations, processes documents with LLMs, automates supplier communications, generates predictions, and produces bid-response documents, each of these workflows has real business consequences when something breaks. This is not just a manual QA role. You’ll write code every day, building test frameworks, automating regression suites, hardening CI pipelines, and creating tooling that makes the entire engineering team ship with confidence. You’ll be the person who defines what “tested” means for this product and builds the infrastructure to enforce it. The platform operates under SOC 2 controls for security, availability, and confidentiality. Test coverage isn’t optional, it’s part of how we demonstrate that changes are controlled and verified before they reach production.
Designing and implementing a comprehensive test strategy across a multi-service architecture , defining the right mix of unit, integration, contract, and end-to-end tests for each service boundaryBuilding and maintaining test frameworks in TypeScript (Jest) and potentially Go as the platform evolves , including shared test utilities, fixtures, factories, and assertion helpersEstablishing test data management patterns, database seeding, tenant-scoped test isolation, deterministic fixture generation, and teardown strategies that allow parallel test executionDefining and enforcing coverage standards, not vanity metrics, but meaningful coverage of critical business logic, data integrity paths, and failure modes
Building automated API test suites for a NestJS backend with 50+ modules, validating request/response contracts, authorization boundaries, tenant isolation, error handling, and pagination behaviorTesting multi-tenant data isolation rigorously, verifying that row-level security policies, tenant-scoped queries, and cross-tenant access controls work correctly under every code pathWriting integration tests for async workflows that span multiple services via message queues, verifying that jobs enqueued by the API are correctly processed by downstream workers and results are persisted accuratelyTesting LLM-dependent workflows, designing strategies to test AI-powered extraction, prediction, and generation pipelines where outputs are non-deterministic. This includes building mock/stub LLM responses, validating output schema compliance, and testing graceful degradation when LLM calls fail or return unexpected formatsTesting email integration workflows, verifying that inbound email parsing, attachment extraction, outbound RFQ dispatch, and response parsing produce correct results across providers (Gmail, Microsoft Graph)
Writing tests that validate schema migrations, verifying that migrations apply cleanly, preserve existing data, and that rollbacks work when neededTesting row-level security policies directly, asserting that queries executed with one tenant’s context cannot access another tenant’s data, including edge cases around shared resources and cross-tenant relationshipsValidating data consistency across async pipelines, when a bid is ingested, items extracted, prices resolved, and proposals generated across multiple services, the final database state must be correct and completeTesting materialized view refresh behavior and ensuring reporting/dashboard queries return accurate results after data mutations
Integrating test suites into CI/CD pipelines as blocking quality gates, tests must pass before code can be promoted from develop to staging to productionOptimizing test execution time, parallelization, selective test running based on changed files, caching strategies, and identifying flaky tests for quarantine and repairBuilding smoke test and health-check suites that run post-deployment to verify that production services are functioning correctly after each releaseIntegrating with existing CI scanning tools (SAST, dependency scanning, IaC scanning) and ensuring test results are visible and actionable in the deployment workflow
Building end-to-end test suites for critical user workflows in the React frontend , bid ingestion, item management, email automation configuration, proposal generation, and prediction reviewImplementing browser-based testing using Playwright or Cypress, with stable selectors, retry strategies, and clean environment setup/teardownTesting document generation outputs (DOCX, XLSX, CSV, PDF) , verifying that template placeholder substitution produces correct documents with accurate data, proper formatting, and no corruption
Identifying and testing performance-critical paths , large bid list queries, bulk item extraction jobs, concurrent LLM processing, and high-volume email ingestionBuilding load and stress test scenarios for API endpoints and async job processing pipelines to validate system behavior under sustained throughput and burst conditionsTesting failure modes and resilience , message queue unavailability, LLM provider timeouts, database connection exhaustion, and S3 access failures. Verifying that the system degrades gracefully and recovers correctly
CategoryTechnologiesLanguagesTypeScript (primary), Go (emerging)Test FrameworkJest (current), Playwright or Cypress (E2E)BackendNestJS, Prisma ORM, PostgreSQLFrontendReactMessagingRabbitMQ (Amazon MQ)AI / MLOpenAI