About the Role
A1 is building a proactive AI chat app for everyday users to bring intelligence to conversations, errands, organising and workflows. Unlike traditional chat-based applications, our product focuses on achieving high reliability for long-running workflows, persistent context, and real-world task completion. The system must handle multi-step reasoning, interact with external tools, and remain reliable despite non-deterministic model behavior.
We are looking for a Full Stack Engineer - AI Systems to build the product layer that turns these capabilities into usable, production-grade workflows. This includes designing how agents operate, fail, recover, and deliver consistent value to users.
Focus
- Build end-to-end product features across frontend, backend, and AI integrations
- Design agent workflows that handle planning, tool use, failure, and recovery across multiple steps.
- Integrate LLMs, memory, and external tools into systems that behave reliably under real-world conditions
- Design real-time AI interactions with streaming, partial results, and tight latency constraints
- Improve system reliability, observability, and fallback mechanisms
- Collaborate closely with ML, backend, and product teams to ship features end-to-end
- Continuously iterate based on real usage and failure modes
Ideal Experiences
- Strong experience in full stack engineering (frontend + backend)
- Solid understanding of system design and API architecture
- Experience working with LLMs, RAG systems, or AI-powered applications
- Ability to handle ambiguity and make pragmatic engineering decisions
- Strong ownership - able to take features from idea to production
- Comfort working in fast-moving environments with evolving requirements
Outcomes
- Own and ship AI-native product features that move beyond chat into persistent, goal-driven workflows
- Design and deploy agent workflows that reliably complete multi-step tasks across tools and sessions
- Reduce latency and improve responsiveness of AI interactions while maintaining output quality
- Build robust fallback and recovery mechanisms for LLM and tool failures in production environments
- Improve the success rate and reliability of AI-driven workflows through iteration, evaluation, and monitoring
- Establish patterns and abstractions for integrating LLMs, memory, and external tools into scalable product systems
- Contribute to a product experience where AI feels proactive, consistent, and dependable over time
Tech Stack
- Next.js
- Python
- NodeJs
- Pytorch
- OpenAI / Anthropic / open-source LLMs
- SQl & noSQL
- Kubernetes
- Docker