AI / ML / OPS ARCHITECT · MISSISSAUGA, CANADA

Intelligence,
engineered for impact.

I design governed, measurable AI systems that turn complex data into decisions people can trust — from agentic workflows and RAG to enterprise platforms and real-time computer vision.

AI
+
OPS
01 / STRATEGY
02 / SYSTEMS
03 / OUTCOMES
23+years across
~7 industries
14+years building
intelligent systems
33AI & IT
case studies
15roles across
industries

01 / ABOUT / THE PRIME PHILOSOPHY

Engineering intelligence.
Earning trust.

Principal AI Engineer · Forward-Deployed AI Architect

I turn regulated, high-stakes business problems into production AI systems — with the evaluation, observability, and governance for reliable enterprise operations.

Mississauga, Ontario, Canada · Open to Principal / Staff AI Engineer and AI Architect opportunities

P

Production-Ready

Reliable, secure, scalable enterprise AI

R

Rapid

Faster delivery and engineering execution

I

Intelligent

Agentic AI, research-driven innovation

M

Measurable

Quality, performance, trust, business outcomes

E

Efficient

Cost-effective architecture and operations

In practice: evaluation and observability as release gates · guardrails and rollback · PromptOps and reusable agentic patterns · model routing and cost control.

01 / WHAT I DO

Agentic AI & GenAI

Multi-agent orchestration, RAG, MCP/A2A, PromptOps, evaluation, tool calling

02 / WHAT I DO

Production ML & MLOps

Architecture, deployment, tracing, CI/CD, model routing, cost optimization

03 / WHAT I DO

Governance & Trust

Responsible AI, explainability, audit trails, PII, regulated-industry controls

Experience 23+ years across approximately seven industries

Career chronology from the GitHub profile and supplied resume. The resume download continues to use the original supplied resume.

Jan 2026 – present

Principal AI Engineer (R&D)

Independent / freelance, Mississauga

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Jun – Oct 2025

AIMLOps Engineer (R&D)

RadComm Systems, Ontario (during graduate studies)

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Jun 2024 – Jun 2025

Lead AIML Engineer

Omdena (during graduate studies)

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Nov 2021 – May 2024

Senior Manager — Solution Architect, AIMLOps & Data Science

Capgemini, Singapore

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Jan – Apr 2021

Data Analytics & Science Manager — Human Capital

Keppel Corporation

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Jan 2018 – Nov 2020

Senior Data Scientist — Data & AI CoE

Deloitte Consulting

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Jan 2017 – Dec 2017

Data Analytics and Data Scientist

Freelance

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Aug 2016 – Jan 2017

Enterprise Solution Software Engineer (R&D) — Telecom

MyRepublic

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Oct 2012 – Feb 2016

Enterprise Software Developer (R&D) — Banking & Insurance

Lexmark Enterprise Software

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

May 2011 – Sep 2012

Signal Processing Engineer (R&D) — Biomedical

Nitto Denko Asia Technical Centre

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

May 2008 – Apr 2009

Vision Software Engineer (R&D) — Biometrics

NEC Asia Pacific

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

May 2006 – Apr 2008

Research Engineer — Data Storage

DSI–A*STAR

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Oct 2005 – Apr 2006

Design Engineer (R&D) — Communication & Power

Panasonic

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Nov 2004 – Apr 2006

Service & Process Engineer — Semiconductor

ASM

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Aug 2001 – May 2003

R&D Engineer (Assistant Manager) — Power & Communication

Duwon Electronics, Yangon

Role and dates from the GitHub profile or supplied resume. Read the source profile ↗

Education & credentials Engineering, automation, signal processing and DevOps
CredentialInstitutionDetail
MEng, Automation & Smart SystemsMcMaster University, Canada2024–2025 · GPA 3.94/4
Specialist Diploma, DevOps/MLOpsRepublic Polytechnic, Singapore2022–2023 · GPA 3.5/4
Innovation & Design ThinkingNational University of Singapore2023
MSc, Signal ProcessingNanyang Technological University, Singapore2003–2005
BE, ElectronicsYangon Technological University, Myanmar1994–2001 · GPA 4.91/5 · First Class Honours

AWS Certified Cloud Architect · NICF PMP · Dataiku / Databricks · Big Data Analytics with Hadoop & Spark · DevOps CI/CD with Jenkins, Ansible & Kubernetes

Engineering toolbox From model evaluation to cloud infrastructure
AreaTechnologies
AI/ML TechniquesMulti-agent orchestration · agentic RAG · prompt engineering · tool calling · semantic retrieval · model routing · fine-tuning · reinforcement learning · self-supervised learning · transformers · vision transformers · RNNs · CNNs · classification · regression · clustering · NLP
LLMs & Agent FrameworksOpenAI/GPT · Claude · Gemini · Llama · Qwen · LangChain · LangGraph · Deep Agents · OpenRouter · Ollama · Model Context Protocol (MCP)
AI Governance & EvaluationRole-based access control · permission gates · human approval workflows · adversarial review · guardrails · evidence-grounded responses · reproducible benchmarks · quality gates · audit trails
AI/ML Libraries & PlatformsTensorFlow · PyTorch · Keras · NLTK · OpenCV · ONNX Runtime · YOLO · DINO · Databricks · Dataiku
Programming Languages & Application DevelopmentPython · Java · C# · C++ · R · Scala · PySpark · Rust · TypeScript · JavaScript · SQL · FastAPI · React · Next.js · REST APIs · WebSockets · WebAssembly
Data Engineering & GovernanceApache Kafka · Apache Flink · Informatica · Hive · Impala · event-driven pipelines · streaming analytics · data reconciliation · BigID · Collibra
Databases, Retrieval & CachingPostgreSQL · pgvector · Qdrant · MySQL · MongoDB · Redis · vector search · exact, semantic, and prefix caching
Cloud & InfrastructureAWS EC2 · S3 · Lambda · Aurora · Azure Virtual Machines · AKS · Azure Machine Learning · Google Cloud Storage · BigQuery · Colab · Docker · Kubernetes
Observability, Analytics & Cost OptimizationPrometheus · Grafana · Langfuse · MLflow · distributed tracing · cost ledgers · inference budget controls · Tableau · Power BI · Google Analytics
Development & Quality ToolsGit · GitHub Actions · Jenkins · pytest · Ruff · mypy · Postman · SoapUI · Visual Studio · PyCharm · CI/CD workflows across Linux, Windows, and macOS

02 / PUBLIC WORKS / RUNNABLE PROOF

Open the work.
Inspect the thinking.

Runnable projects and architecture references, with direct repository links and the scope behind each reported result.

Metrics below are reported in the profile review dated November 14, 2026. Coverage, test counts and evaluation accuracy measure different things. Repository tests were not rerun for this portfolio update. Read the source profile ↗
ProjectWhat it demonstratesResults & Validation
AgentForge Enterprise AI PlatformGoverned agent runtime: hybrid RAG, multi-provider model routing, tenant entitlements, human approvals, hash-chained audit, and cost controls

README scope: Synthetic-data reference implementation; 96% is combined branch-aware coverage. Production provider, storage and identity replacements are documented.

PROFILE-REPORTED96% combined coverage · 269 tests passed · 100% offline golden-case pass rate (20/20)
Agentic Ops Services PlatformMulti-tenant agentic AI foundation: MCP-governed API boundaries, ReAct/ReWOO orchestration, Python/FastAPI, Angular, graph and hierarchical retrieval, and AWS Terraform definitions

README scope: README reports 132 backend tests and 93% line coverage; the profile reports 136 and 91%. Measurement chronology is unresolved. Terraform is not applied; production-scale testing remains pending.

PROFILE-REPORTED91% backend line coverage · 136 backend tests passed · local validation; production-scale testing pending
Agentic Project Management AdvisorCrewAI + local Ollama agents for planning, estimation, resource allocation, and forecasting

README scope: Current standard-library core reports 142 tests and 100% line/branch coverage. CrewAI/Ollama remains an optional prototype outside CI; the historical planning/compliance percentages are not established by those core tests.

PROFILE-REPORTEDPreviously reported: 42% faster planning · 91% structured-output compliance; supporting local benchmark record not located
Agentic Usage Billing Engine

Included in the XERJ Corpus Hub as a reference for coding agents and developers.

View XERJ Corpus Hub ↗
Event-driven, idempotent usage-based billing: ingest → rate → aggregate → reconcile → MCP; serverless on AWS

README scope: Local proof of concept: 291 gating tests plus 3 optional performance tests. AWS is a target blueprint; the README says its cloud deployment is not yet deployable.

PROFILE-REPORTED100% line and branch coverage · 291 gating + 3 performance tests passed
CognitiveMirror-Jev-MCP

Live on Agensi.io.

View on Agensi.io ↗
Cognitive middleware for personal and enterprise agents: 14 MCP tools, REST API, memory, guardrails, approval workflowsPROFILE-REPORTED100% statement and branch coverage · 62 unit tests passed · live PostgreSQL and container checks recorded
ConversaFlow AITenant-aware conversational runtime: calibrated intent classification, explicit dialogue graphs, typed tools, cited knowledge retrieval, REST/WebSocket APIs, and a TypeScript SDKPROFILE-REPORTED94.44% intent accuracy · 94.50% macro F1 (72 synthetic held-out examples) · 171 tests passed
Enterprise AI Reference ArchitectureVendor-neutral architecture knowledge base: AI/ML lifecycle gates, reference designs, reusable patterns, security controls, standards, and architecture decision recordsPROFILE-REPORTED72 starter-kit smoke tests passed · 171 Markdown files and 62 diagrams checked; architecture guidance, no measured business-impact percentage
Industrial Agent IntelligenceEvidence-grounded industrial agents: facility ontology, policy-checked tools, numeric grounding checks, human escalation, and deterministic behavioural evaluationsPROFILE-REPORTED100% behavioural-case pass rate (22/22) · 79.94% combined coverage · 119 tests passed
Med Foundation ModelsWhole-body MRI research framework: self-supervised learning with MAE, SimCLR, DINO and JEPA; Vision Transformers, multimodal fusion, and downstream fine-tuning

README scope: 132/132 is the reported test pass rate, not code coverage or clinical accuracy.

PROFILE-REPORTED100% recorded test pass rate (132/132); clinical accuracy and downstream model gains not established
MGIIA — Mobility AssistantYOLOv8 + DeepLabV3 + Llama 3.2 Vision for hands-free road-scene understanding

README scope: Current decision layer reports 109 tests and 100% core coverage. The original YOLO/DeepLab/Llama-Vision path is outside CI, with no established model accuracy, latency or safety certification.

PROFILE-REPORTEDPreviously reported: 94.7% detection accuracy · 40% fewer false positives; supporting local benchmark record not located
Quant Pulse ProcessorPython/NumPy quantum toolkit: statevector simulation, five canonical algorithms checked against exact references, and pulse-level transmon modellingPROFILE-REPORTED96% coverage · 100% acceptance pass rate (15/15) · 250 tests passed; simulator validation
Sentry-AIOffline cybersecurity signal-processing demo: telemetry replay, statistical baselines, noisy-OR risk fusion, hysteresis

README scope: Offline synthetic replay demo. The README does not enumerate a test count; the profile reports 7/7. Real-data precision and recall are not evaluated.

PROFILE-REPORTED100% recorded test pass rate (7/7) · two synthetic attack replays reach ALERT; real-data precision/recall not evaluated
TrustMed AI FabricGoverned content and AI reference implementation: evidence-backed claims, approval and expiry controls, provenance, MCP tools, cited retrieval, and hash-chained audit

README scope: Deterministic synthetic reference implementation; reported golden-case results do not establish clinical efficacy or production deployment.

PROFILE-REPORTED100% deterministic golden-case pass rate (30/30), citation validity and refusal accuracy · 90% branch-aware coverage · 96 tests
Explore the 33-case-study archive ↓

03 / PRIVATE WORKS / ENTERPRISE DELIVERY

Complex environments.
Practical outcomes.

Anonymized enterprise engagements from the public profile. Outcomes are self-reported; implementation test gates are identified separately.

01 / ENTERPRISE BRIEF

Healthcare AI

Engagement

Springer Nature case study — Top-tier hospital ED capacity

Reported outcome / implementation checks

30% fewer LWBS · 20% faster door-to-doctor · 14% lower discharged-patient LOS · 2,500–3,000 added visits/year

Discuss this engagement ↗
02 / ENTERPRISE BRIEF

Clinical Decision Intelligence

Engagement

North American clinical organization — QEMIDP / Pulse ED — MIMIC-IV + CXR, LangGraph 6-stage pipeline, 5 guardrail layers, HITL, local-first Ollama

Reported outcome / implementation checks

100% evidence attachment (test gate) · 100% guardrail reason-code traceability (test gate) · 5 guardrail layers · 6-stage pipeline · HITL + audit trace

Discuss this engagement ↗
03 / ENTERPRISE BRIEF

Legacy Modernization

Engagement

Local manufacturing company — Legacy-Lens — deterministic LangGraph supervisor, 4 specialist agents, 4 MCP servers, RAG + citation gate, HITL

Reported outcome / implementation checks

41/41 implementation tests passed · 100% citation-gated answers · 8/8 MCP tools · 3 golden eval cases · end-to-end audit trace

Discuss this engagement ↗
04 / ENTERPRISE BRIEF

Edge AI + Multi-Agent

Engagement

Local SME — FIVIS vehicle inspection — LangGraph multi-agent orchestration, RAG/VectorDB, Rust/WASM, ONNX, NVIDIA Triton

Reported outcome / implementation checks

25× performance · 70–85% lower latency · 40–60% scalability · grounded, auditable decisions

Discuss this engagement ↗
05 / ENTERPRISE BRIEF

Drug Discovery

Engagement

Market leader of the pharmaceutical industry — VAE + GNN + RL molecular pipeline

Reported outcome / implementation checks

90% valid molecules · 67.2% quality lift · 14.7% over baseline

Discuss this engagement ↗
06 / ENTERPRISE BRIEF

Disaster Response AI

Engagement

One of the world's healthcare organizations — AI-driven disaster resource matching; Neo4j knowledge graph, DSPy workflows, entity mapping, web scraping, GCP + AWS Lambda

Reported outcome / implementation checks

80–90% faster matching (weeks → hours) · 50%+ coordination efficiency · 60–70% faster resource discovery · 45–55% less manual coordination

Discuss this engagement ↗
07 / ENTERPRISE BRIEF

Agentic Ops

Engagement

Leading wholesale chain — APAMA project management advisor — CrewAI, Ollama, FastAPI

Reported outcome / implementation checks

42% faster planning · 91% compliance · 88% forecast correlation

Historical profile figures; the supporting local benchmark record was not located. Current deterministic-core tests do not validate these percentages.

Discuss this engagement ↗
08 / ENTERPRISE BRIEF

HR & Finance Analytics

Engagement

Asia top-tier services — governed Azure/Power BI platform

Reported outcome / implementation checks

45–55% reporting visibility · 30–40% forecasting accuracy

Discuss this engagement ↗
09 / ENTERPRISE BRIEF

AI Security

Engagement

Japanese bank — SecureNet AI WatchTower — GenAI data-leak monitoring across text, PDF, image, audio, video; risk scoring, RBAC, dashboards

Reported outcome / implementation checks

100% conversation risk scoring · 100% of 5 data types monitored · 3-role RBAC · 4 dashboards · 0% source data retained

Discuss this engagement ↗
10 / ENTERPRISE BRIEF

Data Management & Governance

Engagement

Asia top-tier bank — Project Orion and CATA — real-time and batch data ingestion with CDC/SCD2; Hadoop/Spark/Yarn, Airflow, S3, Collibra, Flink, PySpark, AWS/GCP

Reported outcome / implementation checks

40–50% better ingestion scalability · 45–55% less manual ingestion · 35–45% better data quality · 30–40% faster lake/warehouse availability · 50%+ analytics readiness

Discuss this engagement ↗
11 / ENTERPRISE BRIEF

Enterprise MLOps

Engagement

Top-five financial institution — AIML Workbench / SWOOSH-FRIDAY — Azure ML, Kubeflow, MLflow, DVC, Jenkins, Kubernetes, TruEra

Reported outcome / implementation checks

90% extraction accuracy · 50% faster turnaround · 5+ deep-learning models · RAI-governed

Discuss this engagement ↗
12 / ENTERPRISE BRIEF

Organizational Analytics

Engagement

Big Four consulting firm — ONA, workforce optimization

Reported outcome / implementation checks

50%+ org visibility · 30–40% lower key-person risk

Discuss this engagement ↗
13 / ENTERPRISE BRIEF

AML & Compliance

Engagement

World top-tier financial institution — Real-time screening, entity matching, alert workflow

Reported outcome / implementation checks

30–40% fewer false positives · 35–45% better screening accuracy

Discuss this engagement ↗
14 / ENTERPRISE BRIEF

Workforce Analytics

Problem Statements

  • Organizations needed a data-driven workforce planning platform to align business strategy with future talent demand.
  • Job roles, skills, regional demand, disrupted work, replacement work, and combinational work were difficult to map manually.
  • Workforce decisions lacked strong people-data insights for attracting, developing, retaining, and redeploying talent.
  • Talent planning required better visibility into job fit, manager alignment, team structure, culture, and future disruption risks.

Tech Stacks

  • Data Preparation & Processing
  • EDA & Hypothesis Testing
  • Feature Engineering
  • NLP: Doc2Vec, Bag of Words, GloVe
  • Crosswalk Algorithm
  • Cosine Similarity
  • Classification: Logistic, SVC, Random Forest
  • Clustering: KMeans
  • AWS, PostgreSQL, Tableau, Excel
  • Python, SQL, Bitbucket, Trello, Agile Scrum

Solutions

  • Built a workforce planning analytics platform with pre-processing, modelling, and insights-reporting layers.
  • Developed ontology, Sankey, scenario planning, and work-disruption models to map future workforce changes.
  • Applied NLP, similarity matching, classification, and clustering to connect jobs, skills, regions, and talent needs.
  • Generated business insights to support talent attraction, development, retention, redeployment, and workforce transformation.
  • Delivered dashboards and reports using Tableau, Excel, PostgreSQL, and AWS-based analytics workflows.

Outcomes

  • Improved workforce planning accuracy by 35–45% through data-driven job and skill mapping.
  • Reduced manual workforce analysis effort by 40–50% using automated NLP, clustering, and similarity models.
  • Increased talent strategy visibility by 50%+ through regional, country-level, and role-based insights.
  • Improved scenario-planning efficiency by 30–40% for disrupted, replacement, and combinational work analysis.
  • Strengthened talent optimization decisions by 35%+ by aligning business context, people data, and workforce strategy.

Reported outcomes supplied by Kyaw; not independently benchmarked in this portfolio update.

Discuss this engagement ↗

Private repositories and client documents are not published here. Read the source profile ↗

04 / OPEN-SOURCE WORKS / COMMUNITY

Contribute in the open.
Improve together.

Upstream contributions, active proposals and ongoing community work. Fork creation and open pull requests are distinguished from merged contributions.

ProjectContribution status
DeepEval

Three contributions by MrKyaw, verified 2026-11-16:

  • #3390 · feat(metrics): add deterministic production metrics to community Open · awaiting review
  • #3411 · Feat/community/eval gap metrics Open · awaiting review
  • #3653 · Feat/community/agent as a judge Open · awaiting review

Proposed contributions; the listed pull requests have not been merged.

promptfooContribution ongoing — fork created; continuing upstream PR or commit
vLLMContribution ongoing — fork created; continuing upstream PR or commit
REAIn Progress

05 / RESEARCH & SOLUTIONS / PUBLICATIONS

Research with a path
to application.

  1. Can AI Create Emergency-Department Capacity Without Building More Capacity?

    *Springer Nature*, September 12, 2026 · case study + teaching note

    Find publication on Google Scholar ↗
  2. An MCP-Driven Multi-Agent Enterprise Marketplace Platform Architecture for Myanmar Secondhand and Local Fashion Merchants

    *International Journal of Computer Applications*, August 20, 2026

    Find publication on Google Scholar ↗
  3. Dynamic Functional Connectivity Patterns in Resting-State EEG for Classifying Learning Strategies

    *International Journal of Computer Applications*, December 18, 2025

    Find publication on Google Scholar ↗
  4. Local Pattern Descriptor-based EEG Classification During Mental Arithmetic Tasks: A Comparative Analysis of Machine Learning Models

    *International Journal of Computer Applications*, November 18, 2025

    Find publication on Google Scholar ↗
  5. Intelligent Movement-Controlled Brain-Computer Interface System Based on EEG Signal

    *River Publishers* book chapter, May 25, 2025

    Find publication on Google Scholar ↗

Submitted / under review

Leading solutions in progress

Academic service: Project Supervision · Research Reviewer, Digital Health (July 2026–present)

Additional context from the original portfolio

For hiring teams

Architecture and hands-on delivery across AI, cloud, data and enterprise systems. Explore the career timeline and generate a resume for your role.

For academic collaborators

Clinical multi-agent systems, EEG classification, governed AI and signal-processing security.

For business partners

Evaluate the product portfolio and reported outcomes. Contact Kyaw to discuss a specific problem or product opportunity.

Additional research listings

    Professional contributions from the original resume

      06 / RECOGNITIONS / ARCHITECTURE · COLLABORATION · TEACHING

      The work matters.
      So do the people.

      Perspectives on architecture, technical leadership, mentorship and teaching.

      architectural backbone for our GenAI initiatives

      Highlights architecture leadership across multi-agent systems, LLM applications and platform-independent enterprise delivery.

      Koko MaungSenior Technical Architect · Munich Re
      excels in mentoring junior team members

      Recognizes practical AI proof-of-concept development and mentorship alongside AI/ML and generative AI expertise.

      Saikiran KannanAsia Lead · Data & AI Strategy
      clear, easily digestible explanations

      Highlights the ability to explain complex engineering concepts and support learners through effective teaching.

      Thomas LwinTechnical · ToshibaTec Singapore

      Short excerpts and summaries from recommendations reproduced in the GitHub profile. LinkedIn may require sign-in to read the originals. Read the source profile ↗

      Read LinkedIn recommendations ↗

      THE OPERATING SYSTEM

      Technology is only useful when it earns trust.

      01

      Ground the signal.

      Evidence-first retrieval, clear evaluation gates, and observable decision paths.

      02

      Govern the system.

      Human approval, permissions, audit trails, and responsible AI built in from day one.

      03

      Make impact legible.

      Connect architecture to outcomes: reliability, cost, speed, quality, and adoption.

      A RESUME THAT SPEAKS YOUR LANGUAGE

      Bring the opportunity.
      I’ll bring the signal.

      Enter the job posting link to check its requirements against Kyaw’s source resume. Organization name is optional. We select relevant experience and refine supported wording for the opportunity, then prepare your two-page PDF with original facts and employment dates preserved. You can contact Kyaw for further review.

      ◉Relevant source evidence · Original facts preserved · No external AI

      The public posting is read automatically and matched locally against the stored resume. No external AI or email service is used. The contact link opens your email application; you choose whether to send it. Resumes are processed in memory. Job details and processing metrics are recorded in private operational logs; full resumes and job descriptions are not logged.

      OPEN TO THE RIGHT PROBLEM

      Let’s build something
      worth shipping.