
Can a 1 GB VPS serve a million WordPress requests a minute?
I load-tested WordPress on a 1 vCPU, 1 GB VPS. Local benchmarks said 1 million requests a minute; real Loader.io traffic peaked near 300,000.

“Production AI systems need real engineering, evals, and spend caps.”Aris Setiawan, Founder & Owner

AI Development Agency
for production software,
agents, and AI search
MadeByAris is an AI software development agency owned by Aris Setiawan (13+ years shipping software, First Cursor Ambassador in Indonesia). We build custom AI agents, LLM systems, Next.js & Rust products, and GEO/SEO engineering directly with senior engineers from scoping to launch.
Direct owner reply in 24h
515+ GitHub stars · 100% client IP
Agency Delivery Control Center
Inspect how Aris Setiawan and MadeByAris take AI products from scoping call to production.

Aris Setiawan
13+ YRSFirst Cursor Ambassador in Indonesia · Creator of RankMySEO. Founder-led architecture on every build.
SPEC TURNAROUND
48 hours
DELIVERABLE
Fixed milestone plan
Model trade-offs, latency targets, data boundaries, and expected monthly cloud or API spend documented in 48 hours.
$ madebyaris spec --scope production-ai --budget capped
Partnering with engineering leaders across Europe (EU & UK), Singapore, and Indonesia.
Take an AI pilot to production, or ship custom agents, RAG search, and MCP tools directly inside your repo.
Built with evals, structured logging, fallback routing, and hard spend caps.
Batch Scale
1,000+ / run
Safety Gate
CI Evals
Budget
Hard Cap
Buyers hire MadeByAris because prototypes are easy and production reliability is hard. Inspect real metrics shipped under client traffic and public code.
13+ yrs
Production engineering
Founder-led studio since 2015
515+ ★
Public AI, Rust & SEO code
nca, Spec Kit, rankmyseo/skills
700K → 1.8M
Client search impressions
Per 28 days + RankMySEO in-house
100% IP
Client code ownership
NDA, evals & spend caps built in
Live Global Studio Hours
Direct sync window for Indonesia, Singapore, EU, and UK teams
--:--
Sidoarjo / Jakarta
Studio desk (09:00 to 17:00)
--:--
Singapore
Full-day sync
--:--
Amsterdam / Berlin
Morning window
--:--
London
Morning window
Pick the track that matches your roadmap. Every engagement starts with a written architecture plan and fixed deliverables before any code is billed.
From $2,000
2 to 4 weeks · Fixed scope
Best for taking a stalled AI pilot live or shipping one high-impact AI agent, RAG search, or MCP integration into an existing product.
Custom milestone quote
4 to 12 weeks · Founder-led
End-to-end engineering for AI SaaS platforms, custom web or Rust apps, and Next.js + GEO/SEO rebuilds led directly by Aris plus vetted specialists.
Monthly retainer
Ongoing · EU / UK / SG / ID
Add senior AI architecture capacity directly inside your Slack and GitHub, or level up your engineers with repo-specific Cursor rules and AI PR review.
Written spec before code
Milestones, risks, and expected cloud bills agreed upfront.
Evals & spend caps built in
CI regression gates and hard token budgets on every LLM feature.
100% IP & NDA ready
Your repo, your cloud account, zero proprietary lock-in.
Founder-led engineering
Direct architecture and code review from scoping call to production deploy.
Client work stays under NDA, so our public proof is real production code you can audit on GitHub before booking a scoping call.
Your team bought Cursor seats and the PRs got messier. We set up project rules and an AI review habit on one real repo, then help your engineers adopt it. English or Indonesian.
Clear answers on ownership, pricing, AI safety controls, and timezone collaboration before we jump on a call.
Ask Aris directlyMost agencies pitch you with a senior partner and hand the code to junior developers. MadeByAris is founder-led: Aris Setiawan (13+ years shipping software, First Cursor Ambassador in Indonesia) designs the architecture, leads every repository, and reviews every pull request alongside vetted senior specialists when a project needs parallel tracks.
Meet the founder and review track recordFocused AI production sprints (taking an AI pilot live, shipping an agent, RAG pipeline, or MCP tool with evals and spend caps) start from $2,000 and run 2 to 4 weeks. Full AI SaaS or product builds run 4 to 12 weeks with fixed written milestones. After a 20-minute scoping call, you receive a written architecture plan and quote within 48 hours.
Read what drives project costEvery production AI system ships with golden eval test cases running in CI, structured trace logging, fallback model routing, and hard per-tenant or daily token spend ceilings. We work under NDA, use zero-data-retention API configurations where required, and transfer 100% of IP and repository ownership to your team from day one.
See our production AI engineering processYes. We operate from Indonesia (WIB / UTC+7), giving full-day overlap with Singapore and Indonesia plus direct morning overlap (08:00 to 12:00 CET / 07:00 to 11:00 UK) with European engineering teams. We join your Slack, Linear, and GitHub workflows directly in English or Bahasa Indonesia.
Send a URL, a stalled AI pilot, or two sentences about your stack. Aris reads every inquiry directly, replies within 24 hours, and sends a written architecture plan before any code.
AI agents, Cursor workflows, Next.js, headless WordPress, and technical SEO, written from production systems we ship.

I load-tested WordPress on a 1 vCPU, 1 GB VPS. Local benchmarks said 1 million requests a minute; real Loader.io traffic peaked near 300,000.

Fastify atau Express untuk API Node.js baru? Perbandingan validasi, logging, struktur, dan kecepatan, dengan benchmark saya dan kapan memilih masing-masing.

Clean code adalah kode yang mudah dibaca dan diubah. Lihat tanda kode berantakan, contoh sebelum dan sesudah, dan cara menerapkannya di tim dan AI agent.