Staff AI Agent Engineer

Description
Emergence AI is building next-generation agentic AI systems that move beyond code generation to provable task completion, verification, and long-horizon autonomy. Our Platform focuses on automating complex, mission-critical data workflows, from ingestion and transformation through analysis, decisioning, and action.
We work with customers in semiconductors, life sciences, and other data-intensive industries to turn fragmented, high-stakes data into trustworthy, continuously-operating AI Agents.
We're seeking a Staff/Principal Agentic AI Engineer to define and own the production architecture for our autonomous agent platform. You'll build reliable, safe, and scalable systems that power complex agentic workflows, combining deep AI/ML expertise with systems engineering to create production-grade enterprise agents.
What You'll Do
Build Production AI Agent Systems

Design and deploy end-to-end autonomous AI systems including agent runtimes, memory systems, orchestration, tool integration, and evaluation pipelines
Implement multi-agent orchestration patterns using frameworks like LangGraph, AutoGen, or custom solutions
Build agent coordination patterns for multi-step workflows with self-learning and continuous improvement
Agent Governance & Safety

Embed security, guardrails, sandbox isolation, auditability, and policy enforcement into agent runtimes
Build HITL (Human-in-the-Loop) frameworks with approval workflows and escalation policies
Architect audit trails, decision reasoning logs, compliance tracking, and replay capabilities
Observability & Evaluation

Implement distributed tracing across agent workflows, tool invocations, and LLM calls
Build monitoring dashboards for agent performance, decision quality, and anomaly detection
Design evaluation frameworks and metrics for reliability, latency, cost, and quality assessment
Prompt Engineering & Context Management

Architect centralized prompt repository with versioning, templating, and A/B testing
Build observability systems tracking token usage, quality metrics, and performance
Design context engineering pipelines including RAG architectures
Model Development & ML Pipelines

Train and fine-tune models across supervised, unsupervised, and reinforcement learning paradigms
Build reproducible training pipelines with experiment tracking and hyperparameter optimization
Implement feedback loops connecting user corrections to agent self-improvement
Compensation Range: $200,000 - $230,000 USD
Requirements
Required Qualifications
Systems & Agentic AI Expertise

10+ years building large-scale distributed systems with hands-on Python, Pyspark coding
5+ years production AI/ML systems, with 1+ years on agentic AI architectures
Proven track record building autonomous agents including multi-agent orchestration, tool frameworks, and evaluation infrastructure
Hands-on experience with agent frameworks: LangChain, LlamaIndex, or custom implementations
Built governance and safety systems for autonomous agents in production
Production Engineering

Expert Python and async programming (FastAPI, asyncio, concurrent systems)
Experience with tool-calling LLMs and agentic protocols (MCP or similar)
Production experience with LLM observability and distributed tracing
Deep understanding of multi-agent systems, prompt engineering, and context management
Event-driven architectures and real-time data pipelines
ML & Data Science

Strong ML/statistics background with model evaluation, experimentation, and A/B testing
Experience with model training, fine-tuning, hyperparameter optimization, and feature engineering
Built data pipelines for training, labeling workflows, and feedback loops
Time-series analysis, anomaly detection, or process monitoring experience
Nice to Have

Deep tech domain experience: manufacturing, industrial systems, healthcare, finance, or scientific/engineering applications
Experience translating domain expert knowledge into agent guardrails
Graph databases or knowledge graph reasoning
Preferred To Have

PhD in Statistics, Computer science and relevant field
Experience as a Data Scientist and strong data experience
Why Join Us?

Build real AI agents, not just dashboards. You’ll work on agents that plan, reason, and act across real production environments — not toy demos — with a direct line to customer impact
Work at the intersection of agentic AI and semiconductors. Help some of the world’s most advanced semiconductor companies flag issues earlier, accelerate root-cause analysis, and optimize yield and quality by giving their engineers powerful autonomous assistants
Own problems end-to-end. You’ll have the autonomy to take problems from messy data and ambiguous requirements all the way through modeling, deployment, and iteration in production
Join a small, senior team. Collaborate with experienced engineers, researchers, and product leaders who are deeply invested in pushing the state of the art in agentic AI and shipping it responsibly.
Shape the platform. Your work will directly influence how our agentic data platform evolves — the types of agents we build, the data abstractions we expose, and the capabilities we deliver to customers
If you’re excited about building AI agents that operate on real, high-stakes data — and want to help redefine how semiconductor companies understand and act on their data — we’d love to talk.
Emergence Benefits

Comprehensive health & wellness benefits, including medical, dental, vision, HSA, and FSA options
Voluntary insurance offerings for additional coverage
401(k) plan with company match
Flexible time off and company holidays to support rest and recharge
Flexible remote work environment
One-time home office setup stipend
Ongoing monthly stipend to support your work-from-home needs