Job Title
- Founding Engineer (Full-Stack / AI Platform)
- Technical Lead (AI Product)
- Lead Engineer (Flutter Web + Firebase + LLM)
Job Summary
We're building EmoTree / EaseAI an emotional support and self-awareness product (not therapy, not medical, not emergency services). Our core experience is conversation + structured reflection: users chat in different modes and receive summaries they can revisit.
We're looking for a hands-on Founding Engineer / Technical Lead who can own delivery end-to-end: stabilize the MVP loop, ship improvements fast, and build the technical foundation for scalable retrieval (vector search) and reliable, cost-aware AI experiences.
You'll work closely with the founder/PM, design, and growth, and you'll have meaningful influence over architecture, quality standards, and technical roadmap.
What success looks like
30 days
- Stabilize the MVP loop: auth/token, streaming chat, session persistence, manual summary generation, and basic observability.
- Establish a practical release & rollback routine and a QA feedback loop.
60 days
- Improve reliability and performance (caching/queueing strategy where needed).
- Set up a maintainable prompt/model iteration workflow tied to measurable quality signals.
90 days
- Prepare for scale: clear SLA targets, concurrency strategy (queue + caching), and retrieval foundation (vector DB / hybrid retrieval approach) aligned with product needs.
Responsibilities
- Own MVP reliability and delivery: authentication flows, streaming chat stability, session persistence, and summary generation.
- Collaborate with product/design to turn user flows into shippable, testable increments (prioritize usable over perfect).
- Design and maintain a scalable retrieval approach (vector search; optionally hybrid retrieval) that supports personalization and reflection features.
- Build performance and reliability mechanisms: caching, queueing, rate limiting, and monitoring; define practical SLA targets.
- Establish engineering practices that keep shipping smooth: versioning, QA workflow, release/rollback, incident triage, and documentation/runbooks.
- Make cost-aware decisions for AI components (model selection trade-offs, prompt iteration workflow, evaluation signals).
Qualifications (Must-have)
- Strong software engineering fundamentals and ownership mindset; comfortable leading from ambiguous requirements to shipped outcomes.
- Experience shipping and maintaining production web apps (end-to-end), including debugging and operational awareness.
- Working knowledge of at least one of the following stacks and ability to collaborate across the rest:
- Flutter (Web) and modern front-end patterns, or
- Firebase / GCP (auth, hosting, database, security rules), or
- Backend systems for streaming responses (SSE/WebSocket), session storage, and API design.
- Familiarity with AI product integration patterns (prompt iteration, evaluation basics, cost/performance trade-offs).
- Clear written and verbal communication; able to document decisions and create runbooks.
Preferred (Nice-to-have)
- Hands-on experience with vector databases and retrieval systems (embeddings, indexing, recall/precision trade-offs).
- Experience with LLM fine-tuning or training (even if limited), or strong understanding of model behavior and failure modes.
- Experience designing high-concurrency systems (queueing, caching, backpressure) and setting pragmatic SLAs.
- Prior experience in early-stage startups, shipping MVPs, and iterating quickly with a small team.
- Familiarity with privacy-by-design and secure data handling practices.
Work style / Location
- Location: Taiwan preferred; open to remote within compatible time zones.
- Work mode: Hybrid/Remote (to be discussed).
- Start: ASAP.
