Skip to content
View aayostem's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report aayostem

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
aayostem/README.md

Ayobami Adejumo — Platform Engineering · SOC2 · FinOps

I build the infrastructure that lets startups ship fast without breaking trust.


What I do

Most seed-to-Series A companies hit the same wall: they've moved fast, won customers, raised money — and now those enterprise buyers are asking for SOC2 Type II, HIPAA evidence, PCI-DSS controls, or a Kubernetes architecture that doesn't collapse under load.

That's where I come in.

I'm a Senior Platform Engineer and FinOps specialist who sits at the intersection of three disciplines that most engineers treat as separate jobs: cloud infrastructure, security compliance, and cloud cost engineering. My thesis is simple — infrastructure built for speed is not built for trust, and you can't bolt trust on after the fact.

I work with EU and US companies at the seed-to-Series A inflection point through my consultancy Cloudfrugal, and I'm selectively available for founding engineer engagements where the infrastructure layer is a strategic bet, not an afterthought.


Flagship: Financial RAG Agent

github.com/aayostem/financial-rag-agent

A production-grade multi-agent financial intelligence system built end-to-end — from SEC EDGAR ingestion to LLM-powered Q&A — across 14 architectural layers with 106 passing tests.

What makes it production-grade, not a demo:

Layer Technology
Vector search PostgreSQL + pgvector
Caching Redis
LLM compatibility OpenAI · Groq · Azure (swappable)
API FastAPI
Migrations Alembic
Observability Prometheus + Grafana
Security scanning Trivy · gitleaks · pip-audit
CI/CD GitHub Actions (multi-stage)
Containers Docker multi-stage builds
Orchestration EKS on Terragrunt-wrapped Terraform
Networking Cilium CNI
Secrets HCP Vault + Agent Injector
Autoscaling Karpenter
Service mesh Istio
GitOps ArgoCD ApplicationSets

The full 7-phase Kubernetes production deployment spans 6 Terragrunt modules across 3 environments. The staging cluster (financial-rag-staging-cluster, us-east-1) reached Helm revision 7 with the complete LGTM observability stack, Falco runtime security, and OPA/Gatekeeper policy enforcement.

Companion projects in the same mono-repo infrastructure:

  • Drift Sentinel — ML model drift detection with automated alerting pipelines
  • RiskOracle — Quantitative risk analytics for financial institutions

All three share a Transit Gateway-compatible CIDR allocation strategy for future inter-VPC routing.


Technical depth

Platform Engineering    EKS · Karpenter · Cilium · Istio · ArgoCD · Helm · Terragrunt/Terraform
Security & Compliance   SOC2 Type II · HIPAA · PCI-DSS · HCP Vault · Falco · OPA/Gatekeeper
FinOps                  Cloud cost attribution · Kubernetes cost-per-workload · Spot/on-demand strategy
Observability           Prometheus · Grafana · Loki · Tempo (LGTM stack) · OpenTelemetry
Post-Quantum Crypto     ML-DSA-65 (FIPS 204) · ML-KEM-768 (FIPS 203) · Hybrid ECDSA/X25519
Data                    PostgreSQL + pgvector · Redis · Alembic · SEC EDGAR ingestion pipelines
Languages               Python · Bash · HCL · YAML · SQL

What I've shipped for clients

Alva (post-quantum transaction identity) — Designed the full hybrid PQC architecture for a transaction identity startup building on ML-DSA-65 (FIPS 204) + ECDSA for signatures and ML-KEM-768 (FIPS 203) + X25519 for key encapsulation. Produced the 8-week MVP scope, infra cost model (~$224–384/mo), and the founding engineer engagement structure.

LexClaro — Identified three critical compliance gaps (PII scrubbing for GDPR Art.25, customer-managed KMS, document-level audit trail) after reviewing their AWS architecture. Proposed a targeted Gap Sprint to remediate before enterprise client onboarding.


How I engage

For sub-10 companies (pre-product-market-fit, pre-SOC2):

  • On-Call CTO · Pre-Launch Audit · Gap Sprint · First Hire Infrastructure Review

For 10+ companies (post-Series A, scaling platform):

  • FinOps Principal · Platform Engineering Lead · SOC2 Type II Program Lead · Fractional CTO

Geography: Lagos, Nigeria — serving EU and US clients via Deel / Payoneer / Remote.com


Writing & teaching

I write about the intersection of platform engineering, SOC2, and FinOps as one unified discipline — not three separate jobs.

  • dev.to/aayostem — Long-form technical breakdowns
  • hashnode/aayostem — technical breakdowns
  • Medium/@aayostem — Architecture patterns for seed-to-Series A infra

Five-course portfolio in development:

  1. Financial RAG Agent (26h) — Production ML system on EKS
  2. SOC2 Type II Engineering (20h) — Evidence pipelines, not paperwork
  3. FinOps + IDP Engineering (20h) — Cost attribution at Kubernetes granularity
  4. ML Drift Detection: Drift Sentinel (16h)
  5. Enterprise ML Risk: RiskOracle (18h)

Contact

Consulting: Cloudfrugal Email: aayo.software@gmail.com Available for: Founding engineer roles · Fractional CTO · SOC2 retainers · FinOps audits

If your infrastructure can't pass a vendor security questionnaire, you're leaving enterprise revenue on the table. Let's fix that.

Pinned Loading

  1. drift-sentinel drift-sentinel Public

    ⚙️ Production AWS/GCP/EKS infrastructure platform — Terraform, Kubernetes, Istio mTLS, Vault, multi-cloud IaC, SOC2 audit logging

    Python 5

  2. mlflow/mlflow mlflow/mlflow Public

    The open source AI engineering platform for agents, LLMs, and ML models. MLflow enables teams of all sizes to debug, evaluate, monitor, and optimize production-quality AI applications while control…

    Python 25.2k 5.5k