Open to AI Engineer / AI automation roles

AGE

AI Engineer & Agent Operator

I design prompts, operate agents — loops, cron schedules, and stop conditions — and build apps and websites with them. I also coordinate multi-agent workflows. Python is AI-assisted: I can run and read scripts, and I write them with AI help.

View work Contact X @Daimun220 TODO email TODO LinkedIn URL TODO CV link

Skills

Written honestly. I do not claim production Python or tools I am still learning.

Strong at

  • Prompting — briefs, scoring rubrics, and quality rules for agents
  • Agent operations — loops, cron schedules, and clear stop conditions
  • Building with agents — apps, dashboards, and websites
  • Multi-agent coordination — planner plus executors, shared relays, review passes

Also in the toolbox

  • AI-assisted Python — I can run and read scripts; I write them with AI assistance
  • Human-in-the-loop workflows (no auto-posting unless a human approves)
  • Cloudflare Workers, KV, and D1 in shipped dashboards
  • Browser automation, public APIs, and HMAC-signed REST where a project needed it

Work

Selected projects, grouped by type. Status is labeled as it is: live, running, on demand, paper, shipped, completed, draft, paused, stopped, prototype, or concept. No testimonials, and no numbers I cannot stand behind.

Agents & automation

7 projects

Canborsa quest automation

On demand

Open API integration with HMAC-SHA256 REST auth. Small perpetual round-trips at no more than 5x leverage, closed immediately, with a cost cap. 8 on-chain quests were completed across 2 accounts, and no open positions were left.

What I did

Wired the signed API flow, added the cost cap and immediate flatten, and kept secrets in environment variables.

  • HMAC-SHA256 REST
  • Python (AI-assisted)
  • On-chain quests

OnePump Launch Watch

Paused

A routine that monitors official X accounts @onepumpcc and @Oneswapcc and the launchpad page, then judges whether a mainnet launch is truly live from context — text, media, links, access gates, devnet labels, and waitlists — not from the keyword “live”. If it confirmed a launch, it would set 4 phone alarms about 3 minutes apart and send a chat alert with the source. Change detection used size, MD5, and deploy version. The schedule was every 30 minutes, later every 5 minutes.

The Android phone alarm was tested, including on a locked phone. There were zero false alarms despite teasers, devnet pages, and gated access. Learned limits: automatic phone calls were US-only, timers did not ring, and new alarms needed the mobile app to be connected. Paused on 9 Nov 2026.

What I did

Designed the context-based “is it really live?” checks, the change detection, the alert path, and the Android alarm hook.

  • Scheduled routines
  • X API
  • Change detection
  • Android alarms

Crypto Market Alarm

Paused

Monitors BTC, ETH, SOL, and CC five times a day (07:00, 11:00, 15:00, 19:00, and 22:00 Asia/Jakarta): price, funding, open interest, long/short, news, ETF flows, and 19 curated X accounts. Alerts are written in Indonesian, with practical per-asset guidance and sources. If nothing important is happening, it stays silent.

15 alerts went out in about 3 days (4–6 Nov 2026), including a US midterm dump and rebound, an ETH short squeeze with about $23M liquidated in about 15 minutes, BTC moving back above $80k, and a risk-off stretch with about $1.1–1.24B in 24-hour liquidations. Paused on 9 Nov 2026.

Not financial advice. This is shown as an automation example, not as trading guidance.

What I did

Set the five-times-daily schedule, the data sources, the “stay silent unless it matters” rule, and the Indonesian alert format.

  • Scheduled routines
  • X API
  • CoinGecko
  • Binance Futures API
  • Web search

Automated X content pipeline

Paused

A human-in-the-loop pipeline for crypto ambassador work. It monitors official posts from OneSwap (@Oneswapcc), OnePump (@onepumpcc), and Rocky (@Rocky_exchange) on Canton, plus community mentions. A routine every 3 hours (about 8 runs a day, Asia/Jakarta cron) produced a 3–6 point summary, 2–3 bilingual EN/ID draft posts with source quote links, an AI-generated 16:9 image or a screenshot suggestion, and 3–5 reply suggestions. A separate weekly routine drafted a standalone post.

Quality rules: do not copy source sentences; use numbers, dates, and features only after verification; de-duplicate; flag personal-experience claims for review. Nothing was auto-posted. Ran from early Nov 2026; paused since 9 Nov 2026.

What I did

Designed the brief, the quality rules, and the scheduled research-and-draft loop, with a human reviewing every post before it could go out.

  • X API (public reads)
  • Scheduled routines
  • Research agents
  • AI image generation

OneSwap leaderboard monitor + streak reminder

Paused

Hourly snapshots of the top 10 and of one user account. Point deltas were used to estimate volume per account per hour, along with each trader’s active and off hours. A daily 07:00 report included quiet-liquidity hours and a read-only price-impact table for 2,000 CC across 7 pairs. An hourly streak reminder ran from 07:30 to 18:30, going silent after one confirmation.

About 10 days of hourly data, covering 11 accounts and 7 pairs. Ran 26 Oct–5 Nov 2026; paused on 9 Nov 2026.

What I did

Built the snapshot and CSV pipeline, the daily report, and browser automation with auto re-login via email code (Gmail API).

  • Scheduled routines
  • Browser automation
  • Gmail API
  • Python / CSV

Kabar Lokal

Draft

A local Indonesia / Southeast Asia news evidence agent for Delphi Agent Arena: Evidence Mining (DoraHacks). A Python skeleton is in place, waiting for the API release on 24 Nov 2026.

What I did

Prepared the agent skeleton and the research on local news sources and competition rules.

  • Python (skeleton)
  • Delphi API (planned)
  • Evidence mining

Data & on-chain

4 projects

Canton on-chain transaction investigation

Completed

An anonymous case study. I traced an expired 333 CC swap where the funds arrived about 2 minutes 23 seconds before the deadline, and found a memo with the swap ID written twice, so the payment could not be matched. I wrote an English report for exchange support.

What I did

Followed the on-chain trail, isolated the memo mismatch, and wrote a clear support report. Parties stay anonymous here.

  • On-chain tracing
  • Memo analysis
  • Support report

Canton airdrop research

Completed

Research on Canborsa, Ekiden, and Rocky Exchange — traction, points and seasons, funding, and red flags — plus on-chain verification of a featured app and of validator ownership via explorers, and a set of priority recommendations.

What I did

Compared the three projects, checked claims on-chain where possible, and wrote a priority recommendation.

  • Project research
  • On-chain verification
  • Explorers

XE Network bug bounty audit

Draft, not submitted

Six draft reports (3 High, 3 Medium). A Go proof of concept passes. The reports have not been submitted yet, so vulnerability details are not listed here.

What I did

Reviewed the target, wrote the draft reports, and verified a passing Go proof of concept. Nothing has been filed.

  • Security review
  • Go PoC
  • Bug bounty (draft)

Apps & web

4 projects

AI browser agent extension

Prototype Stopped

A Chrome Manifest V3 extension. A local Qwen model (qwen3:8b via Ollama, CPU-only) plans tasks and routes prompts to a logged-in web session of ChatGPT, Claude, Grok, or Gemini. It has a popup UI. This is a prototype, and work on it has stopped.

What I did

Prototyped the local planner, the prompt routing into already-logged-in web apps, and the popup UI.

  • Chrome MV3
  • Ollama
  • qwen3:8b
  • CPU-only

Tritemius Leaderboard Tracker template

Draft v2

A reusable bot template: a generalized onboarding skill, methodology, and routines for tracking a competition leaderboard. It contains no personal data.

What I did

Turned the leaderboard-tracking method into a reusable template so a new bot can be onboarded without copying personal data.

  • Reusable template
  • Onboarding skill
  • Routines

MLBB TikTok clips

Shipped

70+ second compilation clips for a paid Mobile Legends: Bang Bang TikTok campaign.

What I did

Produced the compilation clips for the paid campaign.

  • TikTok
  • Video
  • Paid campaign

$PARTY meme coin

Concept

A meme coin concept on OnePump. The name and a logo draft are ready. It has not launched.

What I did

Prepared the name and a logo draft. No launch yet.

  • OnePump
  • Brand concept
OnePump on X (@onepumpcc)

Experiments

5 projects

OKX futures paper trading

Paper only

A paper-trading agent on OKX futures with a $100 paper balance. The agent decides its own entries and closes. It keeps a journal and a daily report. Record so far: 1 win, 1 loss (+$0.92). This is paper trading only — the point is the risk controls and the journal, not the P&L.

What I did

Set up the agent loop, the journal, and the daily report, and let the agent take its own paper entries and closes.

  • OKX futures
  • Paper trading
  • Journal
  • Daily report

Rocky Exchange API integration experiment

Stopped

An API integration experiment on Rocky Exchange (Canton). I discovered and verified a Binance-style REST API with HMAC-SHA256 auth, read an account balance (not shown here), and analyzed the CC-USDCX order book and slippage for a 1,000 CC order (about 0.03% spread). I wrote an AI-assisted Python script for taker order execution with anti-wash-trading safeguards. One 20 CC test order settled on-chain. Two 800 CC taker orders matched but failed on-chain settlement, and the funds were returned — a useful lesson in handling settlement failure.

What I did

Verified the signed API, measured book slippage, and wrote a cautious taker script. Account details and balances are not shown.

  • HMAC-SHA256 REST
  • Python (AI-assisted)
  • Order book analysis
  • Settlement handling

RaidLords autoplay agent

Paused

A play-to-earn game on Robinhood Chain. A computer-use browser agent, plus MetaMask on a dedicated wallet, ran an hourly routine: gather resources, upgrade, raid, and claim quests. There was no top-up after about $1 of starting capital. Results: Gold Vault level 8 (300k cap), Keep level 5, trophies from 0 to about 148, and 3 of 4 raids in one session were 3-star wins. Paused since 2 Nov 2026. No wallet or referral link.

What I did

Configured a computer-use browser agent and an hourly loop for the core game actions, on a dedicated wallet with no extra top-up.

  • Computer-use agent
  • Browser automation
  • MetaMask
  • Hourly routine

Magic Chess: Go Go autoplay

Completed

Autoplay through cloud gaming: rule-based bidding each round, lineups, and a per-match self-evaluation loop that fed into the next match, alternating play with me. No win/loss record is listed.

What I did

Wrote the bidding and lineup rules and the after-match review loop, and alternated sessions with the agent.

  • Cloud gaming
  • Rule-based agent
  • Self-evaluation loop

Bull Games operator

Paused

Browser plus MetaMask automation for a game. Currently paused.

What I did

Built the browser and MetaMask automation path for the game loop, then paused the project.

  • Browser automation
  • MetaMask

How I work with agents

I treat agents like operators with a brief, a loop, and a stop condition. I keep a human in the loop for anything that should not be automatic — posting in public, moving funds, or claims that need personal experience.

01

Briefs and prompts

I write a clear brief, a scoring rubric, or a quality rule before a run starts, so the agent knows what “good” looks like and what it must not do.

02

Loops, cron, stop conditions

Work runs on a schedule. Each job has a stop condition — a confirm-then-silence reminder, a cost cap, a flatten rule, or a human review gate — so it does not run on forever.

03

Multi-agent coordination

Larger jobs are split: for example a planner or orchestrator, two executor agents, and a review pass, coordinated through a shared relay rather than one long unstructured chat.

Currently learning

I am learning these in the open. I am not claiming production experience with them yet.

  • n8n

    Just starting. I want to use it for visual automation workflows.

  • Claude Code

    Just starting. I want to get comfortable with it as a coding agent.

  • AI Engineering curriculum

    Currently at phase 14, which covers single agents.

Contact

The fastest way to reach me right now is X. Email, LinkedIn, and a CV are placeholders until I add them.

X @Daimun220 TODO email TODO LinkedIn URL TODO CV link