Back

Daffa Izzuddin

AI Engineer · Malang, Indonesia (UTC+7)

I take models from a training run to something people can actually use: train, serve, monitor, iterate.

daffaizzuddin.w@gmail.com · GitHub · LinkedIn · Writing

Experience

Lead Developer

· RAINS Corp2026-07 to present

Remote · Founding team

An autonomous AI agent that runs alongside clinical and administrative staff and writes the activity reports they would otherwise write by hand.

  • Architected and deployed an autonomous shadow agent that now runs alongside clinical and administrative staff, across physician, healthcare worker and front-office roles.
  • Built the inter-agent query routing that holds a 92% query success rate, and I am benchmarking latency against a sub-5s target.
  • Lead a squad of 2 engineers, ship to staging through CI/CD every day, and steer sprints toward 70%+ milestone completion.
  • Own the outward-facing half of the job too: demo the agent to investors, run the weekly sync and sprint sessions, and open pilot conversations with partner organisations.
  • Shipped the production platform at rainscorp.id and the custom GPT apps that give the agent a natural-language front door.

The agent runs alongside a person's day instead of asking them to change it: it logs the workflow, drafts the report, and the person corrects what it got wrong. The corrections matter, because the agent only earns its place once someone trusts the output enough to stop writing it twice.

python, typescript, nuxt, postgresql, docker

Full-Stack AI Engineer

· Mayfeyr2024-12 to present

Remote · Full-time

Catalogue ingestion, pricing, storefront content, and the AI features on top of them for a multi-brand luxury fashion retailer.

  • Rebuilt catalogue ingestion as a distributed pipeline in Python and RabbitMQ, sized so that onboarding a large supplier catalogue and the daily price sync both run without losing work.
  • Integrated external supplier APIs behind strict retry queues, so a failed sync retries instead of surfacing as a stale price or a negative-margin order.
  • Refactored the pricing engine around DataLoader batching and cut per-product pricing from 100ms to 10ms, a 10x throughput gain.
  • Pulled the hot paths out into Go microservices (pricing, image caching, content generation, PDF rendering) and moved the rest of the background work onto Celery and n8n.
  • Built Mayfeyr Mirror, the virtual try-on engine, from zero to one, plus custom ChatGPT apps that take a conversational shopping query straight to a product.

I also own the multi-server infrastructure and the technical SEO/GEO work that got the full catalogue indexed and crawling cleanly.

python, go, rabbitmq, postgresql, celery, docker

B.Sc. Computer Science

· Universitas Brawijaya2021 to 2025

Education · Malang, Indonesia

Faculty valedictorian, 3.95/4.00 GPA, degree finished in 3.5 years.

  • First author on a Q1 IEEE Access paper for GenBatik, with registered intellectual property and an open-source implementation.
  • Team lead on the Bangkit Academy capstone with Solafune Japan, ranked #1 in the company track and #1 overall across the academy, graduated with distinction.
  • Lab assistant for two years, teaching basic programming, data structures and algorithms, and computer networking at basic and advanced level.
  • Computer vision lab work on retinal blood vessel segmentation and historical manuscript text recognition.

Selected projects

GenBatik

AI, Image Generation, Diffusion, Fullstack

Text-to-image generation for batik patterns: a diffusion model wrapped in a real product, and a paper about it.

https://genbatik.ub.ac.id

SPARSE

AI, Computer Vision, Remote Sensing, GIS

Solar panel detection on satellite imagery, from model to a GIS interface.

IdentifEye

AI, Vision, Embeddings, API

Face-embedding authentication exposed as a small, boring REST API.

TataKata

AI, NLP, Writing

Grammar checking for Bahasa Indonesia, built on EYD and KBBI rules, used by real writers.

https://tata-kata.netlify.app

Quoin

DevTools, Developer Experience, Privacy

Developer utilities that run entirely in the browser: no uploads, no accounts, no telemetry.

https://quoin.izzudd.my.id

ALPR Pipeline

AI, Computer Vision, OCR

Automatic license plate recognition, built from scratch for a national data competition.

AutoMatrix

Data, Publishing, Infrastructure

A dated price index for developer infrastructure, where every number carries the sentence it came from.

https://matrix.invasikode.com

LinkProxy

Go, Streaming, Infrastructure

A streaming gateway that turns any file URL into a stable short link without buffering the file.

Skills

Data & Modeling
Dataset preparation and label sanity checks, Fine-tuning diffusion models (LoRA / SDXL), CNN / embedding models for vision, Evaluation beyond a single accuracy number
Serving & Inference
REST / async job APIs around model inference, Streaming progress to the client (WebSocket), Containerised deploy, GPU vs CPU trade-offs, Latency and cost per request, Message queues and retry policies that survive supplier failures, Batch loading to cut per-product work (DataLoader, 100ms to 10ms), Microservice extraction when one module gets hot
Monitoring & Iteration
Logging inputs/outputs to make failures reproducible, Feedback capture (accepted vs rejected output), Cost and usage tracking per feature, Versioned experiments instead of one-off notebooks, Success rate and latency targets measured, not assumed, CI/CD to staging as the daily default, not a release event
Product Delivery
Nuxt / Vue front ends for data-heavy products, API design and schema work in Postgres, AI features wired into an existing commerce stack, Multi-server infrastructure and technical SEO/GEO, Static-first builds, prerender, and deploy pipelines, Technical writing that explains a system to non-authors, Demoing a working system to investors, and partner conversations for pilots, Running sprint planning and team sync sessions