AI Weekly Malaysia

This week's Lenny's Newsletter covers two main topics: a review of the GLM-5.2 open-source LLM from China, which is competitive with top models on coding and reasoning benchmarks, and a deep dive into how Gusto used Claude Code to rapidly build a new product line, highlighting practical AI agent workflows for engineering teams.

Why: For developers and AI/ML learners, the GLM-5.2 review offers a cost-effective, open-source alternative to proprietary models. For startup founders and engineering leaders, the Gusto case study provides a real-world blueprint for using AI coding agents (like Claude Code) to accelerate product development, which is directly applicable to building in Southeast Asia's cost-sensitive market.

Gusto CTO Eddie Kim shares how a 5-person team shipped a new AI product line in 10 weeks using Claude Code, a permanent Zoom call, and no Figma, Jira, or traditional docs. They relied on AI-generated code, rapid prototyping, and blurred engineering/product roles.

Why: Demonstrates a radical, low-overhead approach to building AI products fast—relevant for Malaysian startups and developers looking to compete globally with lean teams and minimal tooling costs.

The Ministry of Digital has launched a national AI transformation initiative to accelerate AI adoption across Malaysia's public and private sectors, likely involving policy frameworks, infrastructure, and talent programs.

Why: This could unlock government grants, sandboxes, and contracts for local AI builders, while shaping the regulatory environment for AI development and deployment in Malaysia.

Ornith-1.0 is a new open-weight coding model from DeepReinforce, built on Gemma 4 and Qwen 3.5, with sizes up to 397B parameters, achieving top open-source performance on coding benchmarks. It excels at agentic coding with multiple tool calls and runs locally via LM Studio, making it accessible for offline use. Early tests show strong code search and manipulation capabilities.

Why: It provides a free, open-source alternative to commercial coding agents, enabling local, privacy-respecting, and customizable coding assistance for developers and AI agent users.

Hasan.VC concluded its Fund I accelerator with a Demo Day in Bandung, showcasing 20 startups from Cohort 004. Over four cohorts, the programme supported 120 startups and nearly 500 founders across 10 countries, using a people-powered halal venture capital model. The event included live pitches, panels on startup fundability, and a closed-door investment session.

Why: It highlights a growing regional accelerator model with halal VC, offering Malaysian founders insights into what makes early-stage startups fundable and potential funding pathways.

A new report warns that memory (DRAM/NAND) prices will keep rising until 2028 due to a persistent supply-demand gap, with only 60% of demand met by 2027. This extends the ongoing shortage driven by AI, data centers, and device production, directly impacting hardware and cloud costs.

Why: Higher memory costs will inflate cloud bills, server expenses, and device prices for Malaysian startups and developers. Local data center builds and AI/ML projects may face budget overruns, forcing trade-offs in infrastructure planning.

Digital Penang and OSK Ventures have signed a one-year MoU to improve access to venture debt and equity financing for AI, hardtech, and deeptech startups in Penang, targeting the capital gap faced by companies with long development cycles and high capital needs.

Why: This directly addresses a critical bottleneck for Malaysian deep tech founders—access to growth-stage capital. It signals institutional support for startups that don't fit traditional financing models, potentially accelerating commercialization and scaling for AI and hardware ventures in the northern region.

Simon Willison's HTML Table Extractor is a paste-conversion tool that extracts tables from copied rich text and converts them to HTML, Markdown, CSV, TSV, or JSON. It's particularly handy for quickly grabbing structured data from web pages like Wikipedia lists. A related update improves his Rich text to markdown tool with table support.

Why: Saves developers time when scraping or repurposing tabular data from websites, which is common when building dashboards, importing data into databases, or preprocessing for AI/ML. For Malaysian builders, it’s a practical utility for extracting data from local government portals, research reports, or any table-heavy site.

Chamath Palihapitiya raised a $135M Series A for his AI coding startup and stepped in as CEO, signaling continued heavy VC appetite for AI developer tools despite a crowded market.

Why: Validates that AI coding assistants are still a hot funding category. For builders, it means more competition and better tools, but also potential vendor lock-in and pricing shifts as well-funded players scale.

Arena, the widely-used AI model leaderboard, has become a $100M business just months after launching commercial services. The platform helps developers and enterprises benchmark and compare AI models. Its rapid growth signals strong demand for AI evaluation tools.

Why: It shows that model benchmarking is becoming a key business, influencing how developers and companies choose and pay for AI models. The leaderboard’s ratings directly impact purchasing decisions and development workflows.

The JAKIM Sabah website was defaced by attackers claiming affiliation with Anonymous, who allege they stole admin emails. The site was taken offline, and JAKIM has not yet publicly addressed the breach.

Why: Highlights critical cybersecurity vulnerabilities in government digital services, underscoring the need for developers to enforce robust security practices and for startups to consider the reliability of public-sector partners.

NTT Data Malaysia's MD Henrick Choo argues that Malaysian enterprises risk being outcompeted if they don't invest in AI, even if ROI is uncertain. He notes Malaysian firms lag Singapore by ~2 years in AI adoption, but spending continues, with NTT targeting 20% growth. He also stresses Malaysia's need to develop higher-quality AI and cloud engineers to capture global opportunities.

Why: Validates that Malaysian enterprise AI spending is real but cautious. For builders, it signals growing demand for AI integration services and a critical talent gap in AI/cloud engineering that local developers can fill.

DiSCoFormer introduces a single transformer model that jointly learns both the probability density function and the score function (gradient of log-density) across multiple distributions. This unified approach enables tasks like sampling, density evaluation, and out-of-distribution detection without needing separate models.

Why: For AI/ML practitioners, a single model that handles both density estimation and score matching can streamline generative modeling pipelines, reduce maintenance overhead, and potentially improve sample quality and evaluation speed.

A Google AI expert explains the concept of a full-stack AI approach, covering integrated hardware and software layers from TPUs to developer tools. This deep integration optimizes performance, scalability, and efficiency for AI workloads.

Why: Understanding the full AI stack helps builders choose platforms wisely and debug performance issues. It demystifies cloud AI services and guides decisions on whether to use managed solutions or build custom stacks.

Omen AI raised a $31M Series A to build sensors and AI models that monitor liquid coolant quality in data centers, detecting bacterial growth and corrosion before they damage GPU clusters. The system targets the growing liquid-cooled infrastructure powering AI workloads.

Why: As Malaysia and neighbors like Johor rapidly build hyperscale data centers for AI and cloud workloads, liquid cooling is becoming standard. A practical operations tool that prevents downtime and hardware degradation directly impacts cloud cost and compute reliability for builders and startups relying on these facilities.

A Vietnamese repair shop received an NVIDIA RTX 5090 with a completely blown-off 12VHPWR power connector, highlighting ongoing, potentially dangerous power delivery issues with high-end GPUs in the region.

Why: Builders and AI/ML learners using high-power GPUs locally need to be aware of hardware failure risks, especially with 12VHPWR connectors, which can cause downtime, fire hazards, or costly RMA headaches.

OpenAI's new report analyzes how AI will transform jobs across the EU, identifying roles at risk of automation, those poised for growth, and how workflows may change. It provides a data-driven framework for policymakers and businesses to prepare for workforce transitions.

Why: While focused on Europe, the methodology and insights can help Malaysian developers and founders anticipate similar shifts in Southeast Asia, guiding product development and upskilling strategies.

The newly opened LRT3 Shah Alam Line replaces single-trip plastic tokens with QR code-based paper tickets, reducing costs and waste. Rapid KL also confirms plans to introduce an open-loop payment system in the future, allowing riders to pay directly with bank cards or digital wallets without needing a separate transit card.

Why: This signals Malaysia's public transit moving toward open-loop, account-based systems. For developers and founders, it opens opportunities in fintech integrations, digital wallet partnerships, and transit data applications as the underlying payment infrastructure evolves beyond closed-loop stored-value cards.

The LRT3 Shah Alam Line has officially opened, with Rapid KL offering free unlimited rides and parking on the line and its feeder buses from 29 June to 31 July 2026 to encourage adoption.

Why: Improved public transit connectivity between Shah Alam and Klang Valley unlocks new talent pools, reduces commute friction for tech workers, and creates opportunities for transit-based apps, services, and startup offices along the corridor.

Apple may launch a Mac Studio with an M7 Ultra chip and improved thermal design by 2028, as hinted by supply chain leaks. This is a long-term roadmap update with no immediate product impact, but signals a focus on sustained performance for professional workloads.

Why: For Malaysian developers and AI/ML practitioners, high-end Mac Studios are prized for local on-device training, simulation, and creative workflows. A future model with better cooling could handle longer compute-heavy tasks without throttling, influencing hardware investment cycles.

A Chinese company is marketing a handheld gaming PC with specs that rival Valve's Steam Machine at a fraction of the cost, but the claims appear too good to be true, raising scam concerns. The article highlights growing demand for affordable alternatives to Valve's pricey device.

Why: It's a cautionary tale for anyone sourcing hardware from unverified overseas vendors—especially relevant for Malaysian builders eyeing cheap dev kits or gaming rigs. The handheld PC trend could also open niche opportunities for local game developers.

Prime Minister Anwar Ibrahim will unveil Malaysia's redesigned international passport on 30 June at Parliament, featuring 94 security enhancements to combat counterfeiting and forgery. The update focuses on physical document security rather than direct tech infrastructure changes.

Why: For builders working on identity verification, KYC, or travel tech, the new security features may require updates to document scanning or validation systems. However, as it's a traditional passport redesign, the practical impact on most developers and startups is limited.

The LRT3 Shah Alam line began full operations on 29 June 2026, with a month of free rides to boost adoption. The line connects Bandar Utama to Shah Alam, improving commutes for tech workers and potentially making locations along the route more attractive for startups. This infrastructure upgrade could influence where companies set up offices and where talent chooses to live.

Why: Better transit directly impacts the daily lives of developers and startup founders commuting in the Klang Valley, expanding the practical talent pool and lowering friction for in-person collaboration or office relocation decisions.

Simon Willison shares a tiny AppleScript one-liner to count open Safari tabs across all windows: `osascript -e 'tell application "Safari" to count tabs of every window'`. It’s a quick terminal command for macOS users.

Why: Demonstrates how simple automation can surface hidden system data. For developers juggling many tabs, it might spark ideas for personal productivity hacks or browser-based tooling.

A leak claims the PlayStation 6 bill of materials has reached US$960, largely due to memory shortages, potentially forcing Sony to delay the console or price it significantly higher than previous generations.

Why: While directly relevant only to game developers, memory and component price hikes could indirectly impact cloud infrastructure costs, affecting developers and startups reliant on cloud services.