Staff AI Engineer
Velocity Talent
Praca zdalna
- Lokalizacja
- Remote (United States)
- Wynagrodzenie
- USD 230000 per annual
- Godziny pracy
- Full Time
- Opublikowano
- 12 paź 2026
Przed aplikowaniem sprawdź na stronie pracodawcy, czy oferta nadal jest aktualna, i zapoznaj się z pełnymi warunkami.
Opis stanowiska
Interfejs jest w języku polskim. Nazwy stanowisk i opisy pracodawców pozostają w oryginalnym języku publikacji, którym może być angielski.
This is a remote position.
HARD REQUIREMENT — hands-on AWS Bedrock AgentCore experience, OR a multi-agent solution the candidate DEPLOYED to AWS (multiple coordinating agents on real AWS infrastructure; frameworks count when they shipped to AWS; single-agent chatbots, RAG-only, and notebook-only work do NOT clear). US only.Staff AI Engineer for Robots & Pencils. One opening. Fully remote US role with approximately 20% travel (in-person interview and one-week onboarding, expenses covered); start date ASAP. Compensation: up to $230,000. Authorization: must be legally authorized to work in the US; no sponsorship. This is a senior backend/AI engineering role in a mature production AI environment with real enterprise users. The role focuses on Python and Node.js backend engineering, AWS-native serverless architecture, DynamoDB single-table design, and production-grade AI platform engineering: LLM function/tool calling, multi-agent orchestration, vector databases/OpenSearch, RAG, and streaming LLM responses. Responsibilities include building and maintaining backend services, developing APIs, contributing to full-stack integration, supporting workflow orchestration and event-driven processes, improving reliability, troubleshooting production systems, supporting deployment/release processes, participating in code reviews, mentoring, and taking end-to-end ownership of features.
Requirements
THE HARD REQUIREMENT, STATED AGAIN BECAUSE IT IS THE FILTER: hands-on AWS Bedrock AgentCore experience, OR a multi-agent solution the candidate DEPLOYED to AWS (multiple coordinating agents on real AWS infrastructure; frameworks count when they shipped to AWS; single-agent chatbots, RAG-only, and notebook-only work do NOT clear). US only. Qualifying orchestration may use LangGraph, CrewAI, AutoGen, Strands or a custom framework — what matters is that it shipped to AWS. 6+ years professional software engineering experience. Strong Python and Node.js experience. Hands-on experience with LLM function/tool calling, prompt engineering, multi-agent orchestration, vector databases/OpenSearch, embeddings, streaming LLM responses, and RAG. AWS cloud experience including Lambda, DynamoDB single-table design, S3, SQS, EventBridge, and Step Functions. Experience building and maintaining full-stack applications, workflow orchestration tools, API design and backend systems, SQL and NoSQL databases, production environments, unit testing, maintainable test strategies, and Docker. Nice to have: Langfuse, synthetic AI testing, token cost optimization, reasoning effort configuration, and Step Functions for AI workflows.
Originally posted on Himalayas