Skip to content
View sulthannauval's full-sized avatar
  • RantAI
  • Jakarta, Indonesia
  • 03:13 (UTC +07:00)

Highlights

  • Pro

Organizations

@RantAI-dev

Block or report sulthannauval

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
sulthannauval/README.md

Hi, I'm Sulthan

I build AI agents people can hand real work to. Rust and TypeScript, mostly.

# ~/.claw/agents/sulthan.toml

[agent]
name               = "Sulthan Nauval Abdillah"
role               = "Software Engineer @ RantAI"
location           = "Jakarta, UTC+7"
runtime            = ["rust", "typescript"]
hallucination_rate = "low, I check my own outputs"

[agent.behavior]
reply_in_groups = "only when addressed"   # I shipped this rule. I also follow it.

What I'm running right now

RantAIClaw, a Rust runtime for AI employees. I build these parts of it:

  • channels: the agent lives in Telegram, Discord, Slack, WhatsApp and Lark. In group chats it only talks when someone mentions it.
  • memory: every conversation keeps its own notes, so group guests can't read the owner's diary.
  • skills: install, update and remove ClawHub skills from the CLI, TUI or web. A tampered download fails the SHA-256 check and never loads.
  • scheduler: jobs you can pause, edit and rerun, or just ask for one in chat.

Before that

RantAI Agents, an agent platform in TypeScript:

  • a visual workflow builder with 28 node types, so teams automate AI work without code
  • 12 chat artifact types (PDF, PPTX, XLSX, notebooks, React apps), with validators that reject broken LLM output

Side quests

  • SugarCare: a 1D CNN in TensorFlow that tells apart 13 types of diabetes from 10 clinical features, trained on 20K+ patient records. My Bangkit Academy capstone.
  • IEEE AGERS 2025: co-author of a paper on web-based DVB-T2 coverage mapping for Indonesia, using the ITU-R P.1546-6 model.

Long-term memory

I came into software from machine learning: Computer Engineering at Telkom University (GPA 3.87), the Bangkit Academy ML path, and a stack of DeepLearning.AI TensorFlow specializations. I spent two years as a lab assistant teaching Python, C and AI practicums, coordinated 12 assistants, and won "Most Favored Assistant" twice. Teaching is still the fastest way I know to find out whether I actually understand something.

Hot take, open to debate:

  ## indomie-flavors.md
+ Cabe Ijo       // the top pick
+ Goreng         // the OG, still undefeated
+ Ayam Bawang    // light, but it hits the spot
- Soto           // way too salty
- Rendang        // just go buy nasi padang

Stack

Languages and frameworks
Data, infra and tools

Also: Tokio, Axum, SurrealDB, Vercel AI SDK, RAG pipelines, Ollama.

Uptime

GitHub stats GitHub streak

Ping me

#general  recruiter  anyone here building agents in Rust?
#general  recruiter  @sulthan
#general  sulthan    hi! that's me. let's talk:

LinkedIn Email X Instagram Spotify

Pinned Loading

  1. RantAI-dev/RantAI-Agents RantAI-dev/RantAI-Agents Public

    Versatile, Next - Gen Agentic AI Platform

    TypeScript 2

  2. RantAI-dev/RantAIClaw RantAI-dev/RantAIClaw Public

    Multi-Agent Runtime for Production AI Employees

    Rust 2