Tech Stack
ApolloAWSCloudDockerDynamoDBGraphQLJavaScriptMicroservicesNode.jsPostgresRayRedisSQLTerraformTypeScript
About the role
- Design and implement REST/GraphQL APIs in Node.js/TypeScript to serve generative‑AI features such as chat, summarization, and content generation
- Build and maintain AWS‑native architectures using Lambda, API Gateway, ECS/Fargate, DynamoDB, S3, and Step Functions
- Integrate and orchestrate LLM services (Amazon Bedrock, OpenAI, self‑hosted models) and vector databases (Amazon Aurora pgvector, Pinecone, Chroma) to power RAG pipelines
- Create secure, observable, and cost‑efficient infrastructure as code (CDK/Terraform) and automate CI/CD with GitHub Actions or AWS CodePipeline
- Implement monitoring, tracing, and logging (CloudWatch, X‑Ray, OpenTelemetry) to track latency, cost, and output quality of AI endpoints
- Collaborate with ML engineers, product managers, and front‑end teams in agile sprints; participate in design reviews and knowledge‑sharing sessions
- Establish best practices for prompt engineering, model evaluation, and data governance to ensure responsible AI usage
Requirements
- Available working some US hours
- Proficient in Hebrew and English both written and verbal
- 4+ years professional experience building production services with Node.js/TypeScript
- 3+ years hands‑on with AWS, including Lambda, API Gateway, DynamoDB, and at least one container service (ECS, EKS, or Fargate)
- Experience integrating third‑party or cloud‑native LLM services (e.g., Amazon Bedrock, OpenAI API) into production systems
- Strong understanding of RESTful design, GraphQL fundamentals, and event‑driven architectures (SNS/SQS, EventBridge)
- Proficiency with infrastructure‑as‑code (AWS CDK, Terraform, or CloudFormation) and CI/CD pipelines
- Familiarity with secure coding, authentication/authorization patterns (Cognito, OAuth), and data privacy best practices for AI workloads
- AWS certification (Developer, Solutions Architect, or Machine Learning Specialty) [mentioned as having]
- Experience building Retrieval‑Augmented Generation (RAG) systems or knowledge‑base chatbots
- Hands‑on with vector databases such as Pinecone, Chroma, or pgvector on Postgres/Aurora
- Experience with observability tooling (Datadog, New Relic) and cost‑optimization strategies for AI workloads
- Background in microservices, domain‑driven design, or event‑sourcing patterns
- Languages: TypeScript, JavaScript, SQL
- Frameworks & Libraries: Express.js, Fastify, Apollo Server, LangChain‑JS, AWS SDK v3
- Datastores: DynamoDB, Aurora (Postgres + pgvector), Redis, S3
- Infra & DevOps: AWS Lambda, API Gateway, ECS/Fargate, Step Functions, CDK, Terraform, Docker, GitHub Actions
- AI Stack: Amazon Bedrock, OpenAI API, HuggingFace Inference Endpoints, Pinecone, Chroma