Institutional fintech + LLM · US client · de-identified
Regulated-grade LLM infrastructure for an institutional platform
The problem
A US client needed a compliant, institutional-grade technical surface: investor onboarding, dashboards, and AI infrastructure sturdy enough for external security scrutiny.
What I built
- Investor-facing dashboards with a compliance-first onboarding cascade (identity + KYC).
- A multilingual investor concierge chatbot, intent classification, RAG over a curated knowledge base, automated lead capture and document delivery.
- The AWS infrastructure and LLM gateway: a pluggable model layer (Claude, Bedrock, OpenAI, and others), event-driven microservices, Infrastructure-as-Code.
- Helped prepare and harden the infrastructure and deployment readiness for an independent third-party security review.
Architecture
Investor dashboardsKYC onboardingLLM gatewayAWS infrastructure (IaC)
The outcome
A live institutional-grade surface, and infrastructure that stood up to independent external review preparation. Client identities are withheld under NDA. Specifics are available on a call.
What this proves
I build compliance-aware software and regulated-grade LLM infrastructure.
Stack
ReactTypeScriptNode.jsPython/FastAPIAWSTerraformPostgreSQLDifyRAG
Want something like this?
Tell me your most repetitive task, the audit is free.
Book a free AI automation audit