Business Leaders & Managers
Understand what GenAI, LLMs, and AI agents can realistically do for productivity, operations, decision-making, and transformation.
GenAI Training · AI Agent Engineering · Product Advisory
From AI Awareness to
Production AI Capability
I help organizations build practical AI capability — from GenAI awareness and workplace productivity to production-ready AI agents, LLM applications, RAG systems, AI-assisted software development, and responsible AI governance.
25 years in product, software, AI, edge computing, and data center systems architecture. Former Intel Senior Product Manager, Software Architect, and Venture Lead. Former AI Startup CTO/CPO. Current AI Corporate Trainer and Consultant.
25
years in technology, product & engineering leadership
Intel
Former Senior Product Manager, Software Architect & Venture Lead — IoT to Edge Computing
CTO/CPO
Former AI startup Chief Technology & Product Officer
1,000+
tech & non-tech professionals trained, from MNCs to SMEs
HRDC
Certified Trainer — programs claimable under Malaysia's HRD Corp levy scheme
Clients
Trained teams at MNCs, public-listed companies, and local SMEs.
A selection of organizations where I’ve delivered corporate AI training and workshops.
Multinational Corporations
Intel
AMD
Keysight
Motorola Solutions
Altera
Micron
NGK Electronics
Renesas
OSRAM
VAT Group
Indium Corporation
Panasonic
Smith & Nephew
Lexis
Public-Listed Companies & Local SMEs
UWC Berhad
PPH
FM Global
Zhulian
Denim Code
Engtek
Ideal Vision Integration
Volcano Name Plate
SF Metal
NanoSkunk
About
Practical AI capability, built on two decades of real product delivery.
Heng Kar Lau is a Principal AI Trainer, Consultant, and Technology Leader based in Penang, Malaysia. His work helps organizations move beyond AI curiosity into practical adoption and production-ready implementation — covering Generative AI, Large Language Models, AI Agents, AI-assisted software development, LLM engineering, RAG, and AI governance. He has trained more than 1,000 professionals from both tech and non-tech backgrounds, across MNCs, PLCs, and SMEs.
Before focusing on AI training and consulting, he spent more than two decades in software engineering, product management, edge computing, IoT, software testing, data science, and AI platforms. His 24 years at Intel progressed from test engineering and hands-on software architecture, to product and data science leadership for the Internet of Things Group, to Senior Product Management for Intel's edge software platforms — including Intel Developer Cloud for Edge, Intel Edge Software Hub, Intel Edge Software Configurator, and Intel Edge Software Device Qualification Tool. Within that final era, he also led an internal venture under Intel's Disruptive and Innovation Group that secured USD $1M in internal funding. He later served as CTO and CPO of an AI startup, shaping product strategy, engineering processes, AI agent architecture, GenAI solutions, Vision AI SaaS, and on-prem AI products.
Who I help
Built for both business leaders and hands-on engineering teams.
Engineers & Developers
Build practical LLM applications, chatbots, RAG systems, AI agents, API integrations, and AI-assisted software development workflows.
Software Teams & Technical Leaders
Adopt GenAI across the software development lifecycle — requirements, architecture, coding, testing, documentation, and engineering productivity.
SMEs & Corporate Teams
Identify high-value AI use cases, improve workflows, reduce manual effort, and build internal AI capability.
Governance, Risk & Technology Leaders
Understand AI safety, prompt injection, data privacy, governance frameworks, and responsible AI implementation.
What I help organizations do
Training and advisory that connects strategy to production.
AI Training & Capability Building
Design and deliver hands-on AI training programs for technical and non-technical teams.
GenAI Adoption Strategy
Help organizations identify where AI can create real value instead of chasing tools or trends.
AI Agent & Workflow Design
Teach teams how to design AI agents using tools, memory, context, guardrails, evaluation, and orchestration.
LLM Application Development
Guide teams building chatbots, RAG applications, API-based AI tools, local LLM experiments, and domain-specific assistants.
AI for Software Development
Enable engineering teams to use AI across the SDLC: requirements, design, coding, testing, debugging, documentation, DevOps.
AI Safety & Governance
Help teams understand AI risks, prompt injection, privacy, compliance, risk classification, and responsible AI operating models.
Corporate AI training programmes
Faster delivery from the team you have, and AI products you couldn't build before.
Choose what your team needs to achieve, then how deep to go — from one day to five.
Operate Leaner
The lean, AI-native operating model solo builders use to run a one-person company — adapted for every professional on your team.
Unlocking the Power of Generative AI
For managers, HR, finance, marketing, operations and knowledge workers. ChatGPT, Gemini and Copilot, prompt engineering, content creation, data analysis, and responsible AI use.
Practical AI for Engineers
For engineers and technical professionals who want to build with LLMs. LLM fundamentals, API integration, chatbot architecture, local LLMs, prompt design, and lightweight automation.
Claude Cowork for Office Productivity
For any team that runs on documents, spreadsheets and repetitive desk work. Delegating real work to an AI workspace — file and task automation, drafting, research, and reusable workflows.
Ship Faster
Engineers using AI coding agents to write, test and ship production-quality code across real enterprise codebases.
GenAI in Software Development
For engineering teams evaluating AI-assisted development. Where AI fits across the SDLC, the current tooling landscape, AI-assisted coding and testing, and what changes in day-to-day delivery.
GenAI for Software Development: Mastery and Applications
For software engineers and technical leads. Hands-on AI-assisted coding, context engineering, prompt patterns for code, AI-assisted testing, and review workflows.
GenAI for Production Software Development
For engineering teams shipping to production. Spec-driven development, Claude Code and GitHub Copilot in depth, MCP server integration, secure coding, and production-grade AI workflows.
Build AI Products
The capability to build AI products your company couldn't build before — LLM engineering, RAG, fine-tuning and agent systems, owned in-house.
AI Agents Introduction
For technical teams and decision-makers new to agents. What an agent actually is, agent loops, tools and memory, and where agents work well today versus where they still fail.
AI Agents & Protocols Deep Dive
For engineers, architects and consultants building agents. Tools, memory, planning and guardrails, MCP and A2A protocols, LangGraph and LangSmith, and agent evaluation.
Deep Dive LLM Techniques for Engineers and Developers
For developers and AI practitioners building domain-specific applications. RAG architecture, embeddings, vector databases, LLM evaluation, and observability.
From Beginner to Advanced LLM Developer
The full path for technical teams building AI products in-house. LLM fundamentals through RAG, LoRA/QLoRA fine-tuning, local deployment, API hosting, and production evaluation.
Optimize the Plant
LLM, agentic and machine-learning applications built for manufacturing operations and the factory floor.
Foundation and Advanced LLM for Manufacturing Industry
For manufacturing engineers, quality and process teams, and plant technology leaders. LLM fundamentals through advanced applications — process documentation, quality reporting, predictive maintenance workflows, and AI-assisted engineering documentation.
Agentic AI Engineering for Manufacturing
For plant automation and engineering teams. Agent architectures applied to manufacturing — workflow automation, tool integration with plant systems, and guardrails for production environments.
Foundation and Applied Data Analytics & Machine Learning for Manufacturing Industry
For manufacturing engineers, quality assurance teams, and operations leaders. Data analytics fundamentals through applied machine learning — defect detection, yield analysis, predictive quality, and production data pipelines.
Runs across all four outcomes
1 day
AI Safety, Security, Risk & Governance
For risk leaders, compliance professionals, auditors, product managers and technology leaders. AI risk classification, prompt injection, data privacy, compliance frameworks, and organizational AI adoption roadmaps. Delivered standalone, or layered onto any of the four outcomes above.
Open-Source Lab
My repositories, and a curated library — to help you navigate GenAI.
71 public GitHub repos plus 8 published data science reports, organized into two groups: what's actually Done by Me, and a much larger Library of repos I curate and share to help others navigate GenAI — spanning Product & Program Leadership, LLM Foundations, Harness & Loop Engineering, AI Coding, and a Solo-Builder / AI-Native Stack. Only a handful of repos here are genuinely high-impact — they're labelled as such, so you know where to start.
71Repos featured on this page
15Built by me (2 original + 13 coursework)
56Curated & explored (forked)
8Published data science reports
Built by me
15 repos
🏆 Done by Me
The AI PDM Leadership Council — a hybrid-retrieval RAG app — an early chatbot experiment, and thirteen Johns Hopkins Data Science Specialization projects. Code and analysis I actually wrote, not tutorials followed.
Curated
6 repos
💼 Curated — Business
Product & Program Leadership picks — Lenny Rachitsky's PM canon — plus the Solo-Builder / AI-Native stack for running a one-person AI company. Forked references, not my code.
Curated
50 repos
🗂️ Curated — Technical
The engineering side of the library — LLM Foundations, AI Agent Projects, Context Design, Harness & Loop Engineering, AI Coding, Tooling, and Observability. Forked references, not my code.
Snapshot as of 21 August 2026 · the 71 repos above are what's actually featured on this page — see the complete profile at github.com/hengkar for everything else, including 43 pre-2020 web-development forks (Angular, ecommerce, .NET) kept as archive but not shown here.
🏆 Original Builds — 2 repos, public
Not forked, not tutorials followed — designed and built by me. This is the tier that actually demonstrates capability.
🏆 Flagship · High Impact
Product Leadership
AI PDM Leadership Council
The repo here that's genuinely high-impact: original, technically deep, and real evidence of capability. RAG app querying the work of 8 product leaders (Marty Cagan, Teresa Torres, Julie Zhuo, and more) — hybrid dense+BM25 retrieval, cross-encoder reranking, dynamic few-shot prompting, and citation-grounded, anti-hallucination answers.
ai-pdm-leadership-council ↗
Built by me
Early build
AI Chatbot Experiment
A basic Streamlit + GPT-3.5 template — an early, honest first experiment in conversational AI. Included for completeness, not billed as high-impact.
chatbot ↗📊 Data Science Foundations — 13 repos · 8 published reports
Johns Hopkins Data Science Specialization, 2015–2016 — my own analysis, submitted as coursework and published to RPubs. RPubs retires in June 2027, so the write-ups are mirrored here for the long run.
🏆 Only externally forked repo
Getting & Cleaning Data
Human Activity Recognition
Tidies the UCI Human Activity Recognition dataset — smartphone accelerometer and gyroscope readings across six activities. The only repo in this entire library that a stranger has forked.
Human-Activity-Recognition… ↗
Capstone
NLP
SwiftKey Text Predictor
A next-word predictor built with Katz/Good-Turing back-off smoothing over an n-gram model trained on blogs, news, and tweets. Two RPubs write-ups: a milestone report and the final model documentation.
Data_Science_Capstone-SwiftKey ↗
Coursework
Practical ML
Weight Lifting Form Prediction
Predicts the quality of weight-lifting exercise form from wearable-sensor data (the Quantified Self movement dataset) using a random forest classifier.
Practical_Machine_Learning-Assignment ↗
Coursework
Regression
Motor Trend Transmission Analysis
Quantifies how much manual vs. automatic transmission affects fuel economy in the mtcars dataset, controlling for weight and horsepower via multivariate regression.
Regression_Models-Assignment ↗
Coursework
Inference
Statistical Inference, Two Ways
Does Vitamin C dose affect tooth growth in guinea pigs (hypothesis testing)? Does the Central Limit Theorem hold for averages of exponential random variables (simulation)? Two RPubs reports from one repo.
Statistical_Inference-Assignment ↗
Coursework
Reproducible Research
Storm Data Analysis
Which severe weather event types in the NOAA Storm Database cause the most harm to population health and the economy across the U.S., 1950–2011?
RepData_PeerAssessment2 ↗
ExData_Plotting1
ExData_Plotting2
ProgrammingAssignment2
RepData_PeerAssessment1
courses
datasciencecoursera
datasharing
all 8 reports on RPubs ↗
Also on the shelf: two dozen forked references on time-series forecasting, traffic prediction, feature selection, and ML system design — kept for study, not shown individually here.
💼 Curated — Business — 6 forks, product & company-building
Not my code — the product-leadership and solo-builder-economy side of the library. Clearly labelled as forked, not original.
📋 Product & Program Leadership — Lenny Rachitsky's product management canon
Curated fork
PM Skills
28 AI-Powered PM Skills
28 skills distilled from Lenny's Podcast — turns Claude Code and Cursor into a world-class product-strategy partner.
awesome-pm-skills ↗
Curated fork
Transcript Archive
Lenny's Podcast Transcripts
The full transcript archive of Lenny's Podcast — the raw material behind the PM skills above.
lennys-podcast-transcripts ↗
Curated fork
Starter Pack
Lenny's Newsletter + Podcast Data
A starter pack combining Lenny's Podcast transcripts and Lenny's Newsletter posts, pre-formatted as AI-friendly markdown.
lennys-newsletterpodcastdata ↗🚀 Solo-Builder / AI-Native Stack — Garry Tan's Gbrain and personal Claude Code setup
Curated fork
Garry Tan
Gbrain
Garry Tan's opinionated OpenClaw/Hermes agent brain — his personal AI operating system, forked as a reference for solo-builder tooling.
gbrain ↗
Curated fork
Garry Tan
Gstack
Garry Tan's exact Claude Code setup: 23 opinionated tools acting as CEO, Designer, Eng Manager, Release Manager, Doc Engineer, and QA.
gstack ↗
Curated fork
Consulting Skills
Strategy Skills for Claude
21 standalone Claude skills for strategy and consulting work, grouped into six practical domains.
strategy-skills-for-claude ↗🗂️ Curated — Technical — 50 forks, organized by focus area
Not my code — a deliberate library, grouped by outcome rather than dumped in one list. Each group is anchored on one thinker or ecosystem and clearly labelled as forked, not original.
📚 LLM Foundations — core texts and code for understanding how LLMs actually work
Curated fork
O'Reilly
Hands-On Large Language Models
The official code repo for the O'Reilly book — a hands-on tour of embeddings, transformers, and applied LLM techniques.
hands-on-large-language-models ↗
Curated fork
Build-Along
LLMs From Scratch
Implements a ChatGPT-style LLM in PyTorch from scratch, step by step — tokenization through pretraining.
LLMs-from-scratch ↗🧩 AI Agent Projects & Workflows — a working library of agent use cases, RAG systems, and reference implementations
Curated fork
Use-Case Library
500 AI Agents Projects
500 curated AI agent use cases across industries, each linked to a real open-source implementation.
500-AI-Agents-Projects ↗
Curated fork
Tutorials
GenAI Agents
Tutorials and implementations spanning basic to advanced Generative AI agent techniques.
GenAI_Agents ↗
Curated fork
Deep Dive
AI Engineering Hub
In-depth, code-first tutorials on LLMs, RAG, and real-world AI agent applications.
ai-engineering-hub ↗🧠 Agentic Engineering & Context Design — how to design the context, memory, and workflow an agent runs on
Curated fork
Karpathy
Context Engineering
"Context engineering is the delicate art and science of filling the context window with just the right information for the next step." — a first-principles handbook built on Andrej Karpathy's framing.
context-engineering ↗
Curated fork
Workshop
Designing Real-World AI Agents
Hands-on workshop: build a multi-agent AI system from scratch — a Deep Research Agent and Writing Workflow served as MCP servers.
designing-real-world-ai-agents-workshop ↗
Curated fork
Handbook
Agentic Design Patterns
A hands-on guide to building intelligent systems, based on Antonio Gulli's Agentic Design Patterns.
Agentic_Design_Patterns ↗🔁 Harness, Loop Engineering & RSI — recursive self-improving loops for AI coding agents
Curated fork ★1
Loop
Loop Engineering
Patterns and CLI tools for orchestrating AI coding agents, built on Boris Cherny's "design loops, not prompts" philosophy. My only starred fork.
loop-engineering ↗
Curated fork
Software Factory
Squid
A Claude Code plugin running a software-factory-style pipeline — turns a feature spec into a reviewed PR through 5 agents: PA → SWE → Tester → PR Reviewer → On-Call.
squid ↗
Curated fork
Karpathy · RSI
Autoresearch
Forked directly from karpathy/autoresearch — AI agents running research on single-GPU nanochat training, automatically.
🛠️ AI Coding — Boris Cherny's loop philosophy and software-factory-style agent pipelines
Curated fork
Practice
Claude Code Best Practice
From vibe coding to agentic engineering — practice makes Claude perfect.
claude-code-best-practice ↗
Curated fork
Karpathy
Andrej Karpathy Skills
A CLAUDE.md distilled from Karpathy's observations on where LLM coding agents go wrong.
andrej-karpathy-skills ↗
Curated fork
Anthropic
Skills
Anthropic's public repository for Agent Skills — the reference implementation this whole practice is built on.
skills ↗⚙️ Tooling & Environments — the surrounding stack: prompts, environments, and workflow builders
Curated fork
Prompts
AI Enablement Prompts
A library of prompts for enabling AI adoption across a team.
ai-enablement-prompts ↗
Curated fork
Pipeline
DeepCode
Open Agentic Coding — Paper2Code, Text2Web, and Text2Backend in one pipeline.
deepcode ↗
Curated fork
Workflow Builder
Sim
An open-source AI agent workflow builder — a lightweight, visual way to build and deploy LLM-connected tools.
sim ↗📈 AI Observability, Evaluation & Optimization — knowing whether an agent is actually working
Curated fork
Evaluation
Opik
Debug, evaluate, and monitor LLM applications, RAG systems, and agentic workflows — tracing, automated evaluation, production dashboards.
opik ↗🧩 Also Exploring — a distinct hardware/EDA interest, off the core AI-and-product thesis
Curated fork
Hardware/EDA
Vibe-IC
An AI-native IC design system generating RTL from natural-language intent, with automated verification through tape-out — worth showing range, not central to the main thesis.
vibe-ic ↗Experience that bridges AI, product & enterprise adoption
A 25-year arc from test engineering to agentic AI engineering.
Former Intel (2000–2024)
- Test Engineering & Software Architecture (2000–2014) — progressed from Board and System Test Development Engineer to Software Developer/Architect and Agile Coach/Scrum Master/Software Architect, building test capability for motherboard, telecom, and server board products before moving into hands-on software architecture and agile transformation leadership.
- Product & Data Leadership, IoT Group (2014–2019) — served as Product Manager/Product Architect and Data Science Specialist/Software Architect for Intel's Internet of Things Group, leading Unified Software Test Framework (GIO) product vision and building out Intel's data science methodology and analytics practice.
- Senior Product Management, Edge Software (2019–2024) — Senior Product Manager for Intel Edge Software Hub, Edge Software Configurator, and Intel Developer Cloud for Edge — the class of infrastructure production and edge AI workloads run on today — including an 8-month rotation as Venture Lead for TheEasyData.AI under Intel's Disruptive and Innovation Group, securing USD $1M in internal funding and reaching finalist status in an international innovation competition.
That 24-year arc — testing rigor, systems architecture, data and product leadership, then venture and edge-software leadership — now underpins how he teaches AI evaluation, production-readiness, and responsible deployment.
Former AI Startup CTO & CPO
Led product strategy, engineering development, AI agent framework direction, GenAI product roadmap, Vision AI SaaS, on-prem AI solutions, and engineering workflow transformation.
Principal AI Corporate Trainer (2024–Present)
Delivered hands-on AI training through PSDC, covering entry-level GenAI adoption, practical LLM development, advanced RAG and fine-tuning, AI agents, software development with GenAI, and AI governance.
Technical Thought Leadership
Published papers in AI, software testing, analytics, and engineering systems — with a small, actively curated open-source lab spanning the same range.
Why work with me
Training grounded in real product and engineering leadership.
- Combines AI training with real product and engineering leadership experience.
- Has built and led software platforms, AI products, and enterprise technology initiatives.
- Understands both business adoption and technical implementation.
- Teaches AI using practical labs, real workflows, and business-relevant use cases.
- Helps teams move beyond prompt tips into structured AI capability.
- Covers both productivity-level GenAI and production-grade AI systems.
- Helps organizations think about AI safely, responsibly, and strategically.
Ready when you are
Ready to build practical AI capability in your organization?
Whether your team is just starting with GenAI or preparing to build production AI agents, LLM applications, RAG systems, or AI-assisted software workflows, I can help design the right training and advisory program for your needs.