Kaleidoscope, an Infosys Company is hiring an AI Developer / Agentic AI Engineer in Charlotte, NC to design, build, and operate an LLM-powered agentic AI platform transforming commercial banking customer service. This senior engineering role covers LLM agent orchestration, Retrieval-Augmented Generation (RAG), intent classification, tool/API integrations with CRM and core banking systems, human-in-the-loop workflows, evaluation frameworks, observability, CI/CD for agent configurations, and enterprise-grade security — within a regulated financial services environment serving commercial banking operations. Note: Visa sponsorship is not available for this role.
About the Role — Agentic AI for Commercial Banking
Mission: Build an agentic AI platform to transform commercial banking customer service — from email intake to workflow resolution
Agent Stack: LLM orchestration, planner/supervisor patterns, tool-using agents, routing, guardrails, and RAG grounding
Integrations: CRM, workflow engine, core banking services, case management — secure tool/API calling to downstream systems
Compliance Scope: Enterprise-grade controls, observability, audit logging, PII protections, and human-in-the-loop patterns
Team: Cross-functional with Product, Operations, Technology, and Risk partners — production-ready delivery focused
Career Growth & AI Innovation Impact at Kaleidoscope
Infosys Backing: Build cutting-edge agentic AI at a global technology leader — the innovation scale of Infosys behind every project
Fintech Pioneer: Shape the future of AI-driven commercial banking customer service at one of the most impactful use cases in fintech
Full Stack AI: Own everything from RAG pipeline design to production observability — rare end-to-end agentic AI ownership
Regulated Environment: Build AI systems held to the highest standards of risk control, compliance, and governance in financial services
Position Overview
The AI Developer / Agentic AI Engineer at Kaleidoscope in Charlotte, NC designs and builds LLM-powered agentic workflows for commercial banking customer service — implementing intent classification, retrieval-augmented generation (RAG) for policy grounding, tool-calling integrations with CRM and banking systems, human-in-the-loop approval workflows, hallucination reduction strategies, evaluation and regression testing frameworks, production observability for agents, and CI/CD pipelines for prompt and configuration management — all within a secure, enterprise-grade, regulated financial services environment delivered in close partnership with Product, Technology, Operations, and Risk teams.
Why This Role Matters: As AI Developer at Kaleidoscope, you are building the intelligent infrastructure that will redefine how commercial banking customers get served — turning complex inbound requests into accurately classified, knowledge-grounded, workflow-executed resolutions without unnecessary human intervention, but with exactly the right human oversight at the points that matter. Your RAG pipelines provide the policy grounding that prevents hallucinations, your guardrails protect the bank from AI-driven compliance risk, your observability tooling gives operations teams real-time visibility, and your evaluation frameworks ensure the system keeps improving safely over time — making this one of the most consequential AI engineering roles in US fintech today.
Key Responsibilities
Agentic AI Solution Development
- Build and continuously enhance LLM and agent orchestration using planner/supervisor patterns, tool-using agents, routing logic, and guardrails
- Implement intent classification, information extraction, validation logic, and decision workflows for commercial banking servicing use cases
- Develop secure tool-calling integrations to downstream systems — CRM, workflow engines, core banking services, and case management platforms
- Implement human-in-the-loop workflows covering review, approval, escalation, and override patterns based on confidence and risk thresholds
Knowledge Grounding & Retrieval-Augmented Generation
- Design and implement Retrieval-Augmented Generation (RAG) architectures for policy and procedure grounding and resolution guidance
- Build and maintain knowledge ingestion pipelines with robust refresh and versioning capabilities to keep grounding content current and accurate
- Improve answer quality through advanced chunking strategies, embedding optimisation, re-ranking approaches, and context window management
- Work with vector databases — including Pinecone, Weaviate, Milvus, OpenSearch/Elastic, or pgvector — for efficient semantic retrieval at scale
Quality, Safety & Evaluation Frameworks
- Define and run comprehensive evaluation frameworks — including golden datasets, scenario tests, regression tests, and automated scoring pipelines
- Actively reduce hallucinations and operational risk by implementing prompt policies, output constraints, structured responses, and verification steps
- Partner with Risk and Compliance teams to ensure full traceability, audit logging, and explainability requirements are consistently met
- Develop LLM-as-judge evaluation approaches, rubric scoring, retrieval evaluations, and both offline and online testing methodologies
Production Readiness, Observability & DevOps
- Implement comprehensive observability for production agents — covering latency, cost tracking, tool failure monitoring, drift detection, and escalation rates
- Support CI/CD pipelines for agent prompts and configurations — including versioning, approval workflows, and rollback capability
- Collaborate with platform and security teams on secrets management, access controls, PII protections, and safe cloud deployments
- Deploy and operate AI services on cloud platforms — AWS, Azure, or GCP — applying sound DevOps practices for reliable production systems
Qualifications & Requirements
Required Qualifications
- 4+ years of software engineering experience or equivalent — with strong computer science fundamentals as the foundation
- Hands-on experience building with large language models (LLMs) and modern AI application stack including agents, RAG, and tool/function calling
- Strong proficiency in Python and building backend services and APIs — Python is the core language for this agentic AI role
- Experience with at least one major AI framework: LangChain, LangGraph, LlamaIndex, Semantic Kernel, or a comparable framework
- Experience with vector databases and semantic search — Pinecone, Weaviate, Milvus, OpenSearch, Elastic, or pgvector
- Experience deploying services to cloud platforms (AWS, Azure, or GCP) with solid basic DevOps and infrastructure practices
- Strong understanding of security and privacy principles — PII handling, least privilege access, and audit logging requirements
Preferred Qualifications
- Experience in financial services or other regulated domains — risk controls, compliance requirements, and audit readiness processes
- Experience integrating with enterprise workflow platforms such as ServiceNow, custom workflow engines, or BPM/RPA systems
- Familiarity with model risk management (MRM) governance — model cards, risk assessments, and validation processes
- Experience with messaging and eventing systems (Kafka/SQS), email ingestion pipelines, and document processing workflows
About Kaleidoscope, an Infosys Company
Kaleidoscope is a specialist technology and AI company operating as part of Infosys — one of the world’s largest and most respected global technology service providers. This AI Developer role sits at the forefront of Kaleidoscope’s mission to build production-ready, enterprise-grade agentic AI capabilities for the regulated financial services sector — specifically transforming how commercial banking customers receive service through intelligent, LLM-powered, workflow-aware agents. Charlotte, NC provides an ideal location at the heart of the US banking and financial services technology sector, making this role both professionally significant and strategically positioned for long-term impact. Kaleidoscope is an Equal Opportunity Employer — all qualified applicants are considered regardless of background. Note: Visa sponsorship is not available for this role.
Career Excellence: Join Kaleidoscope in Charlotte and build the agentic AI platform that is transforming commercial banking customer service — designing LLM agents, RAG systems, and enterprise-grade AI workflows at the intersection of the world’s most advanced AI engineering and one of its most compliance-critical industries.
Who Should Apply?
- Agentic AI Engineers: With 4+ years building LLM-powered agents, tool-calling workflows, and RAG systems in production environments
- Python AI Backend Developers: With strong API development skills and hands-on LangChain, LangGraph, or LlamaIndex experience
- Fintech AI Specialists: With experience building AI solutions for commercial banking, financial services, or other regulated industry environments
- RAG & Vector Search Engineers: Skilled in designing knowledge grounding pipelines with Pinecone, Weaviate, OpenSearch, or pgvector
- AI Safety & Evaluation Engineers: Who build evaluation frameworks, implement guardrails, and drive hallucination reduction in production LLM systems
