Senior Staff Software Engineer - Agentic AI

Job Summary

The Senior Staff Software Engineer - Agentic AI will design and deliver core agent capabilities, including policy execution, reasoning, orchestration, and failure handling. They will build production-ready agent runtimes—state and memory systems, planners, and tool-routing components—while ensuring safety, reliability, latency, and cost efficiency. This role will contribute to multi-agent workflows for classification, routing, reconciliation, anomaly detection, and decisioning. The role includes implementing deterministic paths where needed, integrating graph/RAG memory, and establishing strong observability and evaluation pipelines. You will mentor engineers and help uphold high engineering standards across the team.

Essential Job Functions

• Design and enhance multiagent frameworks and LLM-powered autonomous systems using goal schemas, planner-executor patterns, and well-defined policy enforcement points. Contribute to the definition and implementation of agent state and memory architectures (short/long-term, episodic, semantic) with strong PII-minimization practices and appropriate TTL strategies. (20%)
• Implement and refine ReAct, CoT, and ToT reasoning loops, and help standardize their use across agent workflows. Build offline and online evaluation signals using curated golden datasets. Contribute to defining and implementing agent-to-agent communication protocols using frameworks like LangGraph, AutoGen, or CrewAI. (15%)
• Measure and improve agent decision reliability using tools such as LangSmith, Arize Phoenix, and Weights & Biases. Design deterministic tool-chaining patterns with appropriate fallbacks, and implement safeguards including idempotency, retries, and circuit-breaker logic. Contribute to integrating grounding techniques to ensure accurate and reliable agent behavior. (15%)
• Architect and implement safety controls such as scoped tool access, allow/deny actions, validators, red-team hooks, and human-escalation paths. Use AgentSpec to define and apply Policy Enforcement Points (PEPs) that prevent high-risk or irreversible agent actions. (15%)
• Operate and scale production agentic AI platforms on Kubernetes by defining agent-centric SLOs, instrumenting deep telemetry (reasoning quality, hallucination rates, tool latency), and supporting on-call through agent-assisted triage, automated rollback/rollforward, and human-gated controls. (15%)
• Establish enterprise safety guardrails, ethical guidelines, and compliance standards to ensure agents operate within legal, regulatory, and organizational boundaries, especially in high-risk or financial workflows. (10%)
• Partner with Product to shape AI strategy and agent user experiences, and collaborate with data and backend teams to ensure agents access high-quality, real-time data across SQL systems, CRMs, SaaS platforms, APIs, and event streams within existing architectures. (5%)
• Communicate complex agentic concepts—such as CoT, RAG, and safety controls—to nontechnical stakeholders to build trust and organizational readiness. Model resilience around model failures and promote AI-native engineering practices, including prompt design and evaluation techniques, across teams. (5%)

Minimum Qualifications

• Bachelor's Degree in Computer Science or related field of study.
• 12+ years of experience in Software Engineering including 2+ years in AI/ML, LLMs, agent-based systems or large-scale distributed systems (e.g., LLM-powered applications, LangChain, LangGraph, AutoGen).
• 3+ years of experience implementing Infrastructure as Code (Terraform or Ansible).

Preferred Qualifications

• Master's Degree in Computer Science or related field of study.
• 15+ years of experience in Software Engineering including 3+ years of experience in AI/ML, Generative AI, or agentic systems.
• 4+ years of experience in Fintech/payments integrations; PCI and data protection familiarity.
• 4+ years of experience in implementing Infrastructure as Code (Terraform or Ansible).

Skills

• Agent Based Modeling
• Amazon Web Services (AWS)
• Datadog
• OpenAI
• Grafana
• Graph Databases
• Large Language Models (LLMs)
• PostgreSQL
• Prometheus (Software)
• New Relic
• Redis
• Telemetry

Work Environment

• Normal office environment.
• Remote or Hybrid, 3 to 4 days per month required in office if within 60 miles of a Bread Financial location.
• On-call rotation to include evenings or weekends.

Physical and Mental Requirements

• Communicate/Hearing
• Communicate/Talking
• Typing/Writing

Other Duties

This job description is illustrative of the types of duties typically performed by this job. It is not intended to be an exhaustive listing of each and every essential function of the job. Because job content may change from time to time, the Company reserves the right to add and/or delete essential functions from this job at any time.

Salary Range

• $155,600.00 - $333,300.00
• Full Salary Range varies by state (e.g., California: $178,900.00 - $416,600.00)

Benefits

• Medical, prescription drug, dental, vision, and other voluntary benefits including basic and optional life insurance, supplemental medical plans, and short and long-term disability.
• Six weeks of 100% paid parental leave for eligible parents after a 180-day waiting period.
• 401(k) plan with immediate enrollment.
• 11 paid holidays.
• Flexible Time Off (FTO) program.
• Paid Sick and Safe Time (PSST) and Illinois Paid Leave where applicable.
• Company stock purchase plan.
• Annual incentive bonus based on individual and company performance.

About Bread Financial

Bread Financial is a tech-forward financial services company providing personalized payment, lending, and saving solutions to millions of U.S. consumers. The company supports associate wellness with diverse benefits and offers career development opportunities. Bread Financial is an Equal Opportunity Employer and participates in E-Verify. The company complies with ADA and provides accommodations for applicants with disabilities.