Skip to content
View mizbamd's full-sized avatar

Block or report mizbamd

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
mizbamd/README.md

Mizbauddin Mohammad

Director-scope Platform & Enterprise Architecture (TOGAF) — Principal Software Engineer, Sr. Manager II

I lead platform and enterprise architecture programs at Fortune-1 scale — managing staff and solutions architects alongside engineering managers, tech leads, and developers. My organization spans 45+ engineers, architects, and managers; 80+ engineers across eight cross-functional teams; and systems consumed by 300+ enterprise applications at 50K+ TPS and 99.99% uptime. Recent outcomes include $5.1M+ annual cloud savings (FinOps + architecture), $1.2M mainframe retirement, and governed GenAI/agentic platforms in production.

I am a player-coach for architects — I set enterprise architecture direction, ADR standards, and review-board bar; I coach 15+ senior and staff engineers into architecture and leadership roles. The repositories below are an open Enterprise Platform Reference Architecture: runnable ADRs, system design, CI, and one-command local runs — the reference patterns I expect architects on my teams to deliver, published for regulated industries (banking, healthcare, insurance, retail).

Reach me: LinkedIn · Medium

Pinned on profile: payments-modernization-platform · item-cost-ledger-platform · agentic-rag-engine · governed-mcp-gateway · supplier-golden-record-platform · mizbamd (this README)


Leadership scope (Director lens)

Dimension Scale
Architect org Manage staff and solutions architects plus engineering managers and tech leads — not only IC developers
Organization 45+ engineers, architects, and managers; 8 cross-functional teams (~80 engineers)
Talent Coach and develop 15+ senior/staff engineers into architecture and leadership; raise architecture bar across domains
Business outcomes $5.1M+ annual cloud savings · $1.2M mainframe OPEX eliminated · ~50% deployment frequency improvement
Platform scale 50K+ TPS · 99.99% uptime · 300+ consuming applications
Programs Merchandising cost/price/negotiations · supply chain · space · analytics · data & AI platform modernization
Governance ADRs · architecture review boards · SLO/SLI standards · Area Tech Review Council (50+ major initiatives)

Featured work — Enterprise Platform Reference Architecture (Java + Python)

Each repo reframes a domain I owned at scale as a domain-agnostic capability — with ADRs, system-design docs, tests, CI, and Docker Compose. The same patterns map to banking, healthcare, asset management, retail, and product companies.

Repository Stack What it proves
payments-modernization-platform Java / Spring Modernize a legacy core with zero big-bang: Strangler Fig, Anti-Corruption Layer, CDC, CQRS + event sourcing, orchestration SAGA + compensation, canary
agentic-rag-engine Python / FastAPI Production RAG: hybrid retrieval (vector + BM25 + RRF), reranking, grounded answers with guardrails, LangGraph agent, eval harness
governed-mcp-gateway Java + Python Governed agentic AI: MCP servers, policy engine, human-in-the-loop, hash-chained audit, PCI/HIPAA/retail redaction
pricing-orchestration Java / Spring MACH + DDD pricing platform: rules engine, workflow orchestration, choreography SAGA, canary
streaming-lakehouse-platform Python / PySpark Medallion lakehouse on Delta Lake: Kafka + CDC, CQRS read models, feature store, data-quality quarantine

A distributed transaction with compensation is a payment settlement (banking), a claims adjudication (healthcare), a trade/position update (asset management), and a price-change rollout (retail). Same SAGA. Same CQRS. Different nouns.


Retail platform reference — Platforms That Can't Fail (5-repo series)

Five retail subdomains — each repo is one technology pillar, wired as one coherent platform story.

# Repository Stack Capability
1 supplier-golden-record-platform Kafka · Cassandra Multi-source supplier MDM → golden record + CDC
2 location-reference-cache Redis · RabbitMQ Location read tier — read-through cache, event invalidation
3 supplier-negotiation-observability Elasticsearch · Logstash Negotiation observability — correlation IDs, regulated redaction, stalled-deal SLOs
4 microspace-planogram-platform Airflow · Flink · Cassandra Planogram pipelines — item × club × fixture at scale
5 item-cost-ledger-platform Temporal · Kafka · CQRS Item cost ledger (keystone) — event sourcing, GraphQL + REST, OpenTelemetry + Grafana SLOs
flowchart TB
  s1["#1 supplier golden record"] --> kafka[(Kafka)]
  s2["#2 location cache"] --> kafka
  s3["#3 negotiation observability"] --> elk[(ELK)]
  s4["#4 planogram pipelines"] --> cass[(Cassandra)]
  kafka --> s5["#5 item cost ledger\nCQRS + Temporal"]
  cass --> s5
  s5 --> gql["GraphQL + REST read APIs"]
  s5 --> mcp["governed-mcp-gateway\nmerchandising tools"]
Loading

Research & focused labs

Lab What it proves
ai-kafka-api-lakehouse AI + Kafka + APIs + lakehouse — complementary paths, no shadow integration
multitenant-ai-scale-finops Multi-tenant AI — isolation, scale controls, cost per successful transaction
bounded-agentic-orchestration Bounded agentic design — mode selector, policy/HITL/audit, supervisor multi-agent
scalable-enterprise-rag Dual-path enterprise RAG — offline prep + online hybrid, ACL, abstain, ops metrics
enterprise-ai-platform-planes Five-plane enterprise AI architecture — model gateway, facade APIs, control/FinOps
finops-platform-landing-zone Azure landing zone IaC — FinOps tags, budgets, AKS, Event Hubs, Databricks, Cosmos/Redis
cassandra-tombstone-lab IJESAT 2024 research — reproducible tombstone/compaction experiments
structured-streaming-retail-slo Supplier CDC freshness + cost projection lag SLOs
quantum-retail-optimization-lab QUBO shelf allocation vs classical greedy (simulator)

Toolbox

Java · Spring Boot · Kafka · Temporal · Cassandra · Redis · Flink · Airflow · GraphQL · Python · FastAPI · PySpark · Delta Lake · Kubernetes · Docker · Azure · GCP · Terraform
CQRS / Event Sourcing · SAGA · MACH · DDD · Strangler Fig · RAG · LangGraph · MCP · Agentic AI · TOGAF · FinOps

Credentials

  • Microsoft Certified: Agentic AI Business Solutions Architect (2026)
  • Microsoft Certified: Azure AI Engineer Associate (2025)
  • TOGAF Enterprise Architecture, Part 1 & 2 (v10, 2025)
  • Oracle Certified Master, Java EE 6 Enterprise Architect (2012)
  • Google Advanced Data Analytics Professional Certificate (2023)

Writing

  • "Managing Tombstones in Cassandra and Elasticsearch," IJESAT, Vol. 24, Issue 3 (2024) — reproducible lab
  • "Live Supplier Catalog and Pricing for Omnichannel," IJCSMC, Vol. 13, Issue 3 (2024)

Pinned Loading

  1. payments-modernization-platform payments-modernization-platform Public

    Reference implementation: modernize a legacy payments/ledger core with no big-bang cutover — Strangler Fig, Anti-Corruption Layer, CDC, CQRS + event sourcing, orchestration SAGA with compensation, …

    Java 1

  2. agentic-rag-engine agentic-rag-engine Public

    Production-grade hybrid RAG: vector + BM25 + reciprocal rank fusion, reranking, grounded answers with guardrails, a LangGraph agent, and a retrieval eval harness. Python/FastAPI.

    Python

  3. governed-mcp-gateway governed-mcp-gateway Public

    Governed Model Context Protocol (MCP) servers in Java AND Python: policy engine, human-in-the-loop approval for sensitive actions, and a hash-chained audit log.

    Java

  4. pricing-orchestration pricing-orchestration Public

    MACH + DDD pricing platform: pluggable rules engine, approval workflow orchestration, and a choreography SAGA with compensation/rollback. Java/Spring.

    Java

  5. streaming-lakehouse-platform streaming-lakehouse-platform Public

    Medallion lakehouse on Delta Lake: Kafka streaming + CDC ingest, CQRS read models, a feature store, and data-quality quarantine. PySpark.

    Python

  6. supplier-golden-record-platform supplier-golden-record-platform Public

    Supplier MDM golden record from multi-source legacy (mainframe/SQL/DB2) with survivorship, Cassandra serving, and Kafka distribution

    Java 2