Skip to main content
NEUN
Back to Careers

BIO Protocol

AI Engineer

NEW
RemoteFull-timeGlobal
šŸ  Remote
RemoteRemote work position availableActivePosted within the last 30 days

Job Description

[AI-summarized by JobStash]

You will design, build, and scale agent systems for planning, tool use, memory, and context management. You will integrate agents with internal and external tools and data sources, implement safety guardrails and sandboxed execution, and develop evaluation, telemetry, and automated scoring systems. You will instrument data pipelines for fine-tuning and reinforcement learning, profile and optimize performance and reliability, and contribute to platform services, APIs, orchestration, CI/CD, and observability.

Requirements

  • ā—Experience building production software in Python or TypeScript
  • ā—Strong systems and API design skills (FastAPI, gRPC, GraphQL or similar)
  • ā—Proven experience shipping LLM applications or agentic systems including tool use, function calling, retrieval/RAG, structured outputs, evaluation, or observability
  • ā—Familiarity with agent and orchestration frameworks (LangChain, LangGraph, AutoGen, CrewAI, MCP) and vector databases (FAISS, Weaviate, Pinecone)
  • ā—Experience with cloud infrastructure and containers (AWS, GCP, or Azure), Docker, Kubernetes, Terraform, CI/CD, and production telemetry
  • ā—Ability to translate research prototypes into robust, scalable systems
  • ā—Experience with fine-tuning and reinforcement learning (RL, RLAIF, RLHF) ā€” nice to have
  • ā—Familiarity with benchmarks and evaluations and with schema, ontology, and provenance design ā€” nice to have

Responsibilities

  • ā—Build agent capabilities for planning, tool use, memory, and context management and ship them into production
  • ā—Integrate agents with internal and external tools and data sources using robust schemas and safeguards
  • ā—Develop quality and evaluation systems including unit tests, regression tests, scenario benchmarks, telemetry, and automated scoring
  • ā—Collaborate with scientists to analyze failure modes and improve performance
  • ā—Ensure outputs are source-traceable and compliant with provenance standards
  • ā—Implement safety measures, guardrails, and sandboxed execution for risky operations
  • ā—Optimize performance and reliability via profiling, idempotency, retries, rate limiting, and uptime management
  • ā—Instrument data pipelines for supervised fine-tuning and reinforcement learning
  • ā—Contribute to agent platform services, APIs, orchestration, CI/CD, and observability

Tech Stack

AutoGPTreinforcement learningontologyCrewAIGraphQLLangGraphPythonTerraformMCPfine-tuning
Expired
Search