AshiDock ท่าอวกาศครอบครัว Ashiran
กลับท่า

CONTROL TOWER · หอบังคับการ

Ashiran Arayachon

Full-Stack & AI Agent Engineer · Founder, Ashiran

Thai — native; English — professional working; www.ashiran.com

ตูมตาม อชิรัญ (Ashiran Arayachon) ผู้ก่อตั้ง Ashiran — เรียนเขียนโปรแกรมด้วยตัวเอง เป็นเจ้าของและผู้พัฒนาเว็บในครอบครัว Ashiran ได้แก่ ashiran.com, ASHA, AshiranScore และ Guarded Link และทำคอนเทนต์บน YouTube ช่อง Tarm Ashiran กับ LiSMF

Self-taught engineer. Five years building, deploying and operating production software alone — product, backend, database, deployment, and the money it had to earn. The last two years went into LLM systems: a multi-tool agent, retrieval, a multimodal document pipeline on a real-money path, and the evaluation harnesses that decide whether any of it is actually working. Not from a CS programme; from systems running in production, unattended, for years.

HOW I WORK

Distrusts a green test. Verifies fixes by breaking them on purpose and checking the suite goes red. A recent guard passed 8 of 8 until a mutation showed the threshold sat exactly where the bug landed. That habit is why the numbers on this page are measurements, not estimates.

TELEMETRYHeadline metrics

23
registered agent toolsschema-driven dispatcher, 20-round tool loop
20
tool-call rounds per turnagent keeps working without a human in the loop
50s → 20ms
hot query, after trigram/GIN indexingdatabase CPU 98% → 18% on the same workload
89%
prompt-cache hit ratemeasured while a frontier API was in the loop — a conversational turn cost under one cent
130
case compliance guard suiteanti-leak, disclaimer, no-fabrication — run before the agent was allowed near a customer
12
fps screen samplingtranslation overlay frame hub
150ms
polling contracttransparent click-through overlay window
70K+
YouTube subscriberstwo channels, full production
9
parallel Claude Code workstreamslane protocol, shared mailbox across two machines, session-start state checks
5 units / idle poll vs 10K per day
API quota engineeringpolling, not posting, was eating the budget — tuning intervals made full-length streams affordable

SYSTEMSSkill branches

AI / LLM / Agents

2 YRS
  • agent orchestration with multi-tool calling
  • RAG
  • prompt engineering
  • LoRA fine-tuning
  • LLM-as-judge evaluation
  • guardrails
  • multimodal OCR and vision
  • real-time streaming inference
  • model routing between self-hosted and frontier
  • self-hosted serving (llama.cpp, vLLM)
  • 23-tool agent with a 20-round loop
  • vector index over a US-equities research corpus
  • 130-case guard suite
  • multimodal slip pipeline on a live money path

Backend & Data

5 YRS
  • Django
  • REST APIs
  • ETL and data pipelines
  • PostgreSQL schema design, indexing, query optimisation
  • Redis
  • two Django platforms live on Azure with dual PostgreSQL
  • 50s → 20ms query, 98% → 18% DB CPU

Frontend

5 YRS
  • Tailwind
  • component-structured vanilla JS
  • responsive SPA
  • Chart.js dashboards
  • responsive SPA chat client with persisted client state and streaming updates, mobile-first

Honest gap: Built on Tailwind + vanilla JS, not React. The concepts transfer; the framework is the one thing left to learn on the job.

Cloud & DevOps

4 YRS
  • Azure
  • Docker
  • Cloudflare
  • Linux
  • Git
  • 24/7 worker fleet with watchdogs and a status board shared across two machines
  • scheduled jobs, log-driven health checks and alerting — nobody else is on call

Languages

5 YRS
  • Python
  • SQL
  • JavaScript
  • HTML/CSS
  • MQL5

FLIGHT LOGTimeline

  1. 2020 – present

    Founder & Lead Engineer

    independent

    • Built and ran revenue-generating web products end to end — product, backend, databases, data pipelines, deployment, infrastructure and monetisation.
    • Architected a multimodal document-understanding pipeline (payment slips → structured data) on a live money path: QR-authoritative extraction, self-hosted OCR and a local LLM doing the reading end to end, with no paid API anywhere in the path.
    • Shipped a financial assistant with compliance guardrails — anti-leak, disclaimer and no-fabrication checks — validated by a 130-case guard suite before it was allowed near a customer.
    • Cost as a design constraint: work runs on self-hosted inference 24/7 wherever a local model is good enough. While a frontier API was in the loop, prompt caching held an 89% hit rate, putting a conversational turn under a cent.
    • Built a real-time translation overlay used on every stream: a frame hub sampling the screen at 12 fps, subtitle-band detection, OCR, translation and speaker identification by face, rendered into a transparent click-through window polling a 150 ms contract — two machines, one JSON contract between them.
    • Built a YouTube Live chat assistant that answers viewers in Thai during streams. A router sends casual talk to a self-hosted model and factual questions to a search-grounded API, verified against a 22-case table; it recognises returning community members from data already in the message poll, avoiding a partner-gated API entirely; and it is built to refuse moderation actions and defer to a human.
    • Engineered around a hard API quota by measuring where it actually went: polling, not posting, was consuming it — 5 units per idle poll against 10K/day.
    • Built and distributed automated trading systems (MetaTrader / MQL5), with their own free and rental distribution models and licence handling.
  2. 2022 – 2023

    Account Manager

    TMGM (ASIC-regulated broker)

    • Supported Thai and English-speaking clients on account and platform issues; coordinated across teams and external stakeholders in a regulated financial product.
  3. 2015 – present

    Content Creator

    YouTube

    • Two channels — Tarm Ashiran (programming and trading) and LiSMF (gaming) — grown to 70K+ subscribers. Full production: scripting, editing, publishing, and the analytics loop behind what gets made next.
    • The streams are where the AI tools get used in front of an audience. The live-chat assistant and the translation overlay were built because they were needed on air, which is a harder bar than a demo: they either work in front of viewers or they are visibly broken.
  4. 2019 – 2020

    First production build — appointment & queue system

    family salon business

    • A booking and queue-management web system. Self-taught from the ground up; the reason everything after it exists.
  5. 2013 – 2019

    Business owner & project sales engineer

    interior design / EvoTech / Mahajak

    • Ran an interior design and construction business (2018-2019). Before that, project sales engineer in home automation, lighting and AV integration for high-end residential and hospitality projects (EvoTech 2016-2018; Mahajak 2013-2016).

PROJECTSWhat is flying

ashiran.com

trading knowledge site — Forex articles, broker guides and trading tools, built solo in Django

Stream toolkit

real-time on-screen translator and a YouTube Live chat assistant, both used on every stream

Self-hosted agent server

multi-model chat and agent server with 23 tools, running on local GPUs

OPEN SOURCEHangar

TRAININGEducation