Hi Gavin,
This is David from Onefish AI.
If you are interested in the below position, please apply.
Thanks
Job description
AI Fullstack Engineer
Location: On-Site & Hybrid
Job Type: Full-time
Salary: Competitive, starting at $80K, with a potential equity + path to $300K based on contribution to business performance (details to be discussed)
1. About Us
Onefish AI is building a revolutionary offline-first PWA that delivers AI consulting to SMEs—100% functional offline. Our hybrid architecture combines browser-based Python execution with intelligent cloud routing to empower small businesses with game-like AI tools that boost efficiency and cut waste.
We're a fast-paced startup pushing boundaries in multi-agent AI for real-world business optimization.
2. Role Overview
We're hiring a founding AI Fullstack Engineer to build our core hybrid architecture from the ground up. This is a rare blend of:
Cutting-edge R&D (multi-agent AI systems)
Advanced browser/cloud engineering (offline PWAs + smart routing)
Direct SME impact (building solutions for real businesses)
You'll ship: Offline PWA → Smart cloud integration → Multi-agent AI consulting workflows
3. Tech Stack Overview:
BROWSER: Svelte PWA + Pyodide / Rust,C++ / Tensorflow.js (Offline ML)
CLOUD: Go Router → FastAPI → PostgreSQL
4. Key Responsibilities
FRONTEND & PWA (Offline Resilience)
Architect the PWA Foundation: Build the Svelte PWA and configure a robust Service Worker to ensure all essential code and assets (including Pyodide runtime and models) are aggressively cached for a guaranteed offline experience.
Pyodide Integration: Implement Pyodide/Web Workers for executing offline Python AI functions, ensuring a non-blocking, high-fidelity user experience.
Tiered UX Design: Develop the logic for Free (local-only, interpreter-based) and Premium (local/cloud toggle, accelerated) AI execution tiers.
Asset Lifecycle Management: Implement seamless, version-controlled updates and smart caching for large Python packages and serialized model files.
BACKEND & CLOUD (Hybrid Routing)
Intelligent Router Development: Build the lightweight Go (Gin or similar) router responsible for dynamically deciding whether a client request should be executed locally (in Pyodide) or routed to the cloud FastAPI worker.
API Development: Develop highly performant FastAPI endpoints for parallel and large-scale AI cloud processing.
Data Synchronization: Design and implement resilient, conflict-free data sync mechanisms between browser storage (IndexedDB) and the central PostgreSQL database.
DevOps & Infrastructure: Manage Dockerized services and orchestrate production deployments across cloud platforms (e.g., Netlify/Heroku).
AI/ML INTEGRATION
Runtime Portability: Develop single-source Python AI functions capable of execution within both the constrained Pyodide runtime and the native FastAPI environment.
Model Optimization: Employ techniques (quantization, pruning, knowledge distillation) to create highly efficient, lightweight models specifically for offline execution.
Multi-Agent Systems: Design and deploy sophisticated multi-agent LLM systems to automate complex business consulting workflows.
SME DEPLOYMENT
Work directly with small businesses to analyze workflows and build tailored AI solutions.
Manage production deployments (e.g., Netlify/Heroku for PWA, cloud platforms for APIs).
5. Must-Have Technical Skills
Core Stack
Svelte, Pyodide, FastAPI, Go (Gin), PostgreSQL, Docker
Python
Expert. Single-source code for browser + cloud runtimes
Frontend
PWA development, Svelte/TypeScript/Vite, service workers
Backend
FastAPI APIs + basic Go HTTP routing
DevOps
Docker Compose, Netlify/Heroku deployment
AI/ML
Pyodide ML, model optimization, multi-agent LLM systems
Business Tools
Microsoft Office Suite (Excel for workflow analysis, PowerPoint for client presentations, Word for documentation)
6. Preferred Qualifications
Architecture: Proven experience with hybrid cloud-edge applications and data synchronization strategies.
Advanced AI: MS/PhD in CS/AI or equivalent research experience in LLMs or specialized models.
Performance: Experience with highly performant Wasm pipelines using Rust/C++ or deep knowledge of TensorFlow.js/ONNX.js for browser GPU acceleration.
Ecosystem: Familiarity with vector databases (e.g., Pinecone, ChromaDB) or advanced data science tooling.
7. Compensation & Growth
Starting: $80K base + competitive equity (discussed at time of offer)
Upside: $300K+ total comp potential (salary + bonuses + equity + benefits)
Benefits: PTO (other benefits based on company milestone, discussed at interview)
Growth Path: Lead Engineer → CTO trajectory
8. Why Join Us?
Own the architecture as founding engineer
Transform SMEs with genuinely needed AI tools
Unique stack: Pyodide + Go + Svelte + multi-agent LLMs
Prestige: Lead cutting-edge edge-compute AI R&D with real business impact
9. Application Process
Submit: Resume + GitHub + 1-page cover letter + 3 references (1 Character Reference and 2 Technical References)
Introductions + SME Case Study: 1-hour
Technical Screen: 1-hour
Live Coding: 2-hours
Offer: Within 1 week
Apply: jobs@onefishai.com
Subject: "Candidate: AI Fullstack Engineer - [Your Name]"