AI Systems Engineer · Product Builder
Mako IT LabI build software that works beyond the demo. AI on small devices, memory for agents, and the mobile and web products that bring it all together.
01 / Projects
Assistive navigation needs real-time obstacle understanding without a network connection.
YOLO11 / ONNX / Python / Computer Vision
Offline camera-to-detector-to-spatial-reasoning-to-TTS pipeline with desktop and INT8 edge profiles.
Ongoing Mako R&D for safety devices, including YOLO-family model training; FPS figures are documented targets rather than published benchmarks.
Safety devices need modular AI pipelines that fit constrained compute, memory and power budgets.
YOLO / LoRA / MLX / llama.cppSplit vision, safety validation and event-triggered small-model reasoning with LoRA, MLX, llama.cpp and quantized deployment targets.
Active Mako R&D across YOLO-family model training and constrained-device deployment planning.
LLM agents forget useful context or overfill prompts with noisy history.
TypeScript / Embeddings / Vector search / Agents
Semantic memory layer with compression, embeddings and retrieval for long-running agents.
Independently implemented the SimpleMem paper in TypeScript and published it as @sheriax/simplemem; benchmark figures belong to the cited paper.
Documents need searchable, grounded answers instead of generic chatbot responses.
TypeScript / Bun / SQLite / MiniLM
PDF ingestion, chunking, MiniLM embeddings, SQLite vector store and OpenRouter/DeepSeek answer generation.
Built a CLI RAG system with separate ingest and query commands.
Short-form video production has repetitive transcription, analysis and assembly steps.
OpenAI / Whisper / GPT-4 Vision / PythonWhisper transcription, vision analysis and scripted media generation pipeline.
Automated core pieces of a TikTok/Reels/Shorts production workflow.
Financial recovery requires spending context, receipt understanding and practical guidance.
Flutter / Firebase / OpenAI GPT / ML Kit OCR
Flutter app with Firebase, ML Kit OCR, rule-based tracking and OpenAI GPT for receipt scanning, advice, budgets, savings targets and alerts.
Built a production AI finance wellness app for spending tracking and financial recovery.
Users want a companion-like AI product that is simple enough for everyday use.
Flutter / OpenAI / Google AI / Firebase
Mobile AI companion with OpenAI/Google AI integrations and Firebase-backed user flows.
Built and shipped as a personal Android product, later folded into Sheriax and now discontinued.
Language-learning communities need safer content and lightweight AI assistance.
Flutter / AI moderation / Firebase
Mobile learning app with moderation workflows and Firebase product infrastructure.
In-development Sheriax product with Youhana as sole developer.
Teams need local-first whiteboarding with reliable collaboration and sync.
React / TypeScript / WebSocket / Firebase
React canvas workspace with realtime collaboration, cloud sync and offline-first state.
Built a multi-board collaborative whiteboard platform.
A social-commerce product needed to evolve from its original cross-platform foundation into a richer video-first marketplace.
React Native / Flutter / Next.js / AWS Amplify
React Native and Next.js in the Cubanin phase, followed by a full Flutter rewrite with Stripe Connect, S3 and Firebase as Echify.
Served as Tech Lead and Lead Mobile Developer, leading the full React Native-to-Flutter rewrite during the Echify rebrand.
Construction logistics needs offline support, GPS tracking and dispatch visibility.
React Native / Next.js / Google Maps / WebSocketsOffline-first mobile and web stack with maps, realtime tracking and route optimization.
Built logistics workflows around field reliability and realtime coordination.
02 / Current R&D
Mako R&D combining trained vision models, deterministic safety checks, compact language-model reasoning and edge-oriented deployment.
Current R&DCamera input · YOLO-family models · Dataset tooling · Edge inference
Detection validation · Geometry checks · Event thresholds · Structured outputs
Compact LLM · MLX + LoRA · GGUF export · Event-triggered use
Warnings · Guidance · Multilingual speech · System integration
Training, export and integration artifacts are verified. Published throughput figures remain engineering targets until backed by reproducible device benchmarks.
03 / Experience
2024 - Present
2023 - Present
Agentic systems · RAG · Vector search · Prompt engineering · Tool calling
ONNX · YOLO · LoRA · MLX · llama.cpp
Node.js · Hono · PostgreSQL · Firebase · Docker
React · Next.js · React Native · Flutter
AWS · Cloudflare · CI/CD
MacBook M4 · RDK X5 · Qualcomm QCS6490
Cursor · Codex · Claude · OpenAI · VSCode · Docker
Terminal-heavy · Local prototypes · Small-device testing · Fast iteration loops
04 / Open Source
Lifelong memory for LLM agents with semantic compression and vector retrieval.
PackageFlutter plugin for Stripe Connect embedded onboarding, payouts and payments UI.
PackageDart/Flutter translation key generation from JSON files with type-safe output.
PackageCLI for wiping AWS account resources across regions in disposable dev/test accounts.
Package05 / Writing
The real reason your AI prompts fail has nothing to do with prompting techniques — it's about how clearly you communicate as a person. Here's how learning to share goals instead of tasks changed everything for me, both with AI and with people.
AI agents forget everything between sessions. SimpleMem fixes that with semantic lossless compression — a three-stage pipeline that compresses, synthesizes, and retrieves long-term memories with 30× fewer tokens. Here's how @sheriax/simplemem brings this framework to TypeScript and Node.js.
Building a marketplace or platform app? flutter_stripe_connect is an open-source Flutter plugin that lets you embed all 13 Stripe Connect components directly into your Flutter app across iOS, Android, and Web — with just a few lines of code.