Freelance AI Agent Orchestrators in Jakarta
Freelance AI Agent Orchestrators in Jakarta
Sign Up
Post a job
Sign Up
Log In
Filters
2
Projects
People
Kelvin Desman
Jakarta, Indonesia
Mind of an engineer and the heart of an entrepreneur.
$50k+
Earned
1x
Hired
11
Followers
Follow
Message
Mind of an engineer and the heart of an entrepreneur.
1
Most LLM telemetry posts brag about what they monitor. Here's mine — including what I'm still missing. Built this for lokerdollar.com (http://lokerdollar.com), solo, in production: Covered: — p50 / p95 / p99 latency per task — cost per 1k tokens + daily burn + per-task spend — provider success rates + circuit breaker state — cache hit rate, quota / rate limit tracking — model leaderboard scored on live traffic Gaps I'm still chasing: — prompt versioning tied to quality signals — TTFT for streaming UX (matters for chat tasks) — per-user cost attribution — semantic error taxonomy, not just HTTP codes — push-based degradation alerts (today's playbook is reactive) The lesson from running this solo: observability is never "done." It's a backlog that evolves with the product — and the gap list is more honest signal than the green checkmarks. → Open for AI platform / full-stack work. If you want this layer for your team, my DMs are open.
1
65
1
The LokerDollar.com web up to 85% faster in some cities — by moving our code farther from users. Sounds backwards. Here's why it worked. 1. The edge promise is seductive: run your code in 300+ cities, and it's milliseconds from every user. Newark, Sydney, São Paulo — all local. 2. Your code doesn't work alone. It talks to a database. And that database lives in one place. A single page render can mean 6+ sequential trips to it. 3. Put your compute "at the edge," and each of those trips crosses an ocean and comes back. The edge didn't remove the distance. It multiplied it. That's how a Sydney request quietly turned into seconds of pure waiting. The fix was one line of config Cloudflare smart placement stopped stop pinning our code next to the user, and run it next to the database instead. One long trip to reach compute, then all the DB chatter is local and nearly free. The result, measured on our own traffic (P90 — the slowest 1 in 10 requests): • Newark ↓ 85% • Sydney ↓ 82% • Melbourne ↓ 63% • São Paulo ↓ 60% and …12 cities faster overall The lesson I'm taking with me: Latency isn't about distance to the user. It's the number of long trips on the critical path. Put compute where it talks the most — for data-heavy apps, that's your database, not the edge. #WebPerformance #EdgeComputing #Cloudflare #BuildInPublic #SoftwareEngineering
1
1
58
1
Excited to share that Lokerdollar has been accepted into the Cloudflare for Startups program. Grateful for the chance to build something impactful. But the bigger story is what we're building. The world of recruiting and hiring is changing fast. In Indonesia alone, millions of people are losing their jobs — economic headwinds, the tech winter, and more. The core problem is a widening gap between industry and talent. That's why we're building an AI career operating system — an AI that acts as a trusted second brain for landing jobs, improving careers, and handling the manual work. Our platform curates jobs based on your interests and expertise, identifies which roles actually increase your chance of getting hired, and automatically fills out applications so you don't burn hours on paperwork and follow-ups. We're also building trust into the system. Sometimes AI screening flags too aggressively and rejects good candidates before a human ever reviews them. Our system catches this — because AI shouldn't be rejecting people. It should be helping them. We're starting in Indonesia, but the opportunity extends far beyond a single country. South Korea, Taiwan, Germany, Italy, the United Kingdom, and the United States are all facing aging populations and shrinking workforces — and they need new ways to connect with talent. Cloudflare has been instrumental in helping us move from concept to a working product in operation. But our vision is much larger than a curated remote job board. We believe every job seeker should have an AI hiring copilot, get continuous improvement between interviews, and that every company should have greater visibility into how they hire. The future of hiring isn't about replacing humans. It's about giving job seekers superpowers. If you're a job seeker, a recruiter, a fellow startup builder, an investor, or someone passionate about the future of hiring and recruitment — I'd love to connect. #FutureOfWork #CloudflareForStartups #BuildInPublic #CareerTech #StartupIndonesia
1
1
75
1
I didn't believe the TypeScript 7 speed claims. So I benchmarked it on my own production app — not a toy repo. 1,358 source files. 260K+ lines. The live codebase behind lokerdollar.com (http://lokerdollar.com). One change: tsc → tsgo (TS7, rewritten in Go). Same tsconfig, same tree, --noEmit. Cold type-check: 77.1s → 11.6s. 6.6× faster. −85% wall-clock. Then the part most "10× faster" posts skip — I checked whether it still catches bugs. Seeded 4 classic type errors. Byte-identical diagnostics: same codes, same line and column. The speed isn't bought by missing errors. But I'm not putting it in your CI yet — and here's the honest reason: — It's a release candidate — The native binary drops the JS typescript API, so typescript-eslint type-aware rules, ts-morph, ts-node and ts-jest break — Custom transformers and incremental builds are still maturing My call: run it now as a fast local / pre-PR check, and keep tsc as the authoritative CI gate until 7.0 ships stable. You get a 6.6× tighter local loop with zero risk to the pipeline. The ~$190/year in saved runner minutes is rounding error. The real return is nobody waiting on a turtle. #TypeScript #DevTools #WebDev
1
1
67
AI Agent Orchestrator
(1)
Follow
Message
Hafidz Noor
Jakarta, Indonesia
Specialist in Finance App & Dashboard Design
Follow
Message
Specialist in Finance App & Dashboard Design
6
Recruiters don't need more tools. They need tools that actually work together. 🤝 CVs buried in email. Slack blowing up with "what's the status?" Candidate data copy-pasted into yet another spreadsheet. It's exhausting. So I built something about it. Inside Notion. 👇 Introducing the Notion ATS Agent, a custom AI agent that handles your entire recruitment pipeline from the second a candidate hits submit. Here's what's running under the hood: ✅ Parses CVs into clean, structured data automatically 📊 Scores each candidate against the actual job description 🔍 Cross-checks LinkedIn PDFs and flags anything that doesn't add up 🔔 Pings your Slack channel at every key pipeline stage 💬 Moves candidates with a simple conversational command No new tools. No extra subscriptions. Just Notion being wildly underestimated again. Honestly the hardest part? Making it behave with real-world messy inputs. Inconsistent CVs, half-filled profiles, vague job descriptions. That's where most of the iterations went. Try it yourself 👇 🔗 Agent: https://www.notion.so/marketplace/custom-agents/applicant-tracking-system 🎥 Demo: https://www.notion.so/agent/33001f30433880839f720092d9faceb4?source=copy_link
1
6
211
3
Rebuilt my personal site as a timeline of everything I've been up to. Click here 👉 https://hafidznoor.com/
2
3
48
1
Jenius - Digital Banking
1
11
0
Lumina – ATS Job Platform
0
1
AI Agent Orchestrator
(1)
Follow
Message
Explore people
