Staff Engineer / AI Builder - NYC

Robots and Pencils•New York, NY

5d•$165,604 - $228,491

About The Position

Robots & Pencils is seeking a Staff AI Builder to lead the design and delivery of AI/ML systems. The role provides technical leadership on a cross-functional team, defining AI architecture, leading model development and optimization, and resolving challenging integration problems. The work involves in-flight production systems where reliability, security, and scalability are critical. The engineer will establish standards, improve AI engineering practices, and contribute to decisions that shape the long-term scalability of the company's AI work. The position focuses on building agentic and generative AI systems for real-world enterprise workflows. Responsibilities span agentic architectures, retrieval-augmented generation, evaluation and observability, fault tolerance, backend and platform engineering, cloud deployment, and frontend integration. The role also involves partnering with product, design, and delivery leads, mentoring engineers, and owning features and releases through production troubleshooting and hardening. Robots & Pencils describes its work as combining engineering with creativity to create AI systems for people and real workflows. The company states that it ships production-ready AI in 30 to 45 days. The posted salary range is $165,604–$228,491 USD. Eligible employees receive paid time off, medical, dental, and vision insurance, and a 401(k).

Requirements

At least 6 years of professional software engineering experience, including substantial experience shipping GenAI or LLM-powered systems in production.
Hands-on expertise in agentic AI, including reasoning loops, tool or function calling, and multi-agent orchestration, with a considered perspective on when single-agent designs are preferable.
Practical RAG expertise, including chunking strategies, embeddings, vector databases such as OpenSearch, cosine similarity search, and re-ranking.
Experience evaluating and observing LLM systems using golden datasets, LLM-as-judge, RAGAS or comparable metrics, and tracing tools such as LangFuse or LangSmith.
Strong prompt engineering skills, including the ability to write structured production system prompts with meaningful constraints.
Hands-on experience with the AWS GenAI stack, including Bedrock, Agent Core, Lambda, DynamoDB single-table design, S3, SQS, EventBridge, and Step Functions.
Strong Python and Node.js skills, with experience building full-stack applications and RESTful APIs.
Understanding of production reliability patterns for LLM-backed systems, including retries and backoff, circuit breakers, and fallback models.
Experience with Docker and cloud-native deployment.
Ability to own ambiguous, integration-heavy problems and explain technical decisions in depth.

Nice To Haves

Direct experience with Amazon Bedrock Agent Core, including agent registration, MCP tool wiring, memory and session management, and gateway provisioning.
Experience in a regulated or high-stakes domain such as education, healthcare, or financial services.
Familiarity with workflow orchestration tools such as Step Functions, Sequencer, or similar state-machine-based systems.
A perspective on LLM observability tools gained from running them in production.

Responsibilities

Design and build agentic workflows, including reasoning loops, tool and function calling, and orchestration across single- and multi-agent architectures.
Build and maintain RAG pipelines, including chunking, embeddings, OpenSearch vector search, re-ranking, and handling stale or changing knowledge sources.
Integrate with AWS Bedrock and Agent Core, including designing MCP-based tools with precise descriptions.
Write and iterate on production system prompts with structured roles, output constraints, and few-shot examples.
Build evaluation and observability into agent systems using golden datasets, RAGAS-style metrics, LLM-as-judge approaches, and tracing tools such as LangFuse or LangSmith.
Design for LLM failures with retries, backoff and jitter, circuit breakers, fallback models, and clear user-facing degradation behavior.
Build backend services in Python and Node.js, including serverless systems using AWS Lambda and API Gateway.
Design DynamoDB single-table schemas for conversation state, agent memory, and session history.
Support event-driven orchestration using Step Functions, SQS, and EventBridge, with reliable error handling.
Contribute to frontend integration so agentic features work cleanly for end users, and write clean, well-tested code across the stack.
Support deployment, monitoring, and production troubleshooting in cloud-native environments using AWS and Docker.
Contribute informed technical judgment to architecture discussions and articulate the trade-offs among design patterns.
Partner with product, design, and delivery leads across global teams to scope and ship features end-to-end.
Mentor engineers on the pod and improve agentic engineering practices, including the use of AI-assisted development tools such as Claude Code.
Own assigned features and releases end-to-end, including debugging, hardening, and maintaining production reliability.