- Position: Senior Full-Stack & Autonomous AI Systems Engineer
- Type: Rolling Contract
- Location: London (Hybrid)
Role Overview
VersAI is an elite, lean AI shopping ecosystem delivering premium, multimodal conversational experiences for the luxury market. We are proving that a tight team of 3–4 AI-Native engineers can out-deliver a traditional 30-person engineering department.
We are not looking for someone who just writes prompts or relies on basic AI code generation. We are seeking a hardcore, elite Full-Stack Engineer and Architect who is deeply proficient in Spec-Driven Development and capable of building production-grade autonomous AI agents. You will bridge the gap between traditional enterprise software architecture and advanced, real-time AI orchestration, utilizing AI-assisted engineering tools simply as a mechanism to 10x your output velocity.
Core Deliverables & Scope
- Spec-Driven Architecture: Design, maintain, and scale our proprietary Conversational Journey Runtime. You will lead with Spec-Driven Development—defining explicit schemas and interfaces before a single line of execution code is generated.
- Autonomous AI Agent Orchestration: Architect and orchestrate multi-agent swarms (Spec Producers, Reviewers, and Coders) running asynchronously via task backlogs to automate platform features.
- Real-Time Multimodal Streaming: Build and optimize bidirectional streaming pipelines capable of handling low-latency WebSockets and WebRTC connections for live text and PCM audio packets.
- Enterprise Search & Data Infrastructure: Transition legacy data pipelines into greenfield, asynchronous database models native to Google Cloud Platform (GCP), powered by advanced semantic search.
- Continuous Evaluation & CI/CD Guardrails: Architect and maintain a robust Google Cloud Build pipeline that moves beyond simple code compilation. You will design continuous evaluation ("eval") loops that run deterministic and model-based tests against agent shadow deployments, automatically blocking promotions to production if tool-routing, data grounding, or latency thresholds degrade.
Technical Stack & Requirements
1. Primary Full-Stack & Real-Time Engineering (Core Focus)
- Backend: Native fluency in Python (FastAPI/async frameworks), asynchronous PostgreSQL (asyncpg), and Redis caching layers.
- Frontend: Strong experience integrating modern React/Vite frontends with complex backend schemas.
- Streaming & Networking: Deep, hands-on experience with WebRTC, WebSockets, and handling low-latency bidirectional streams (audio/text data packets).
2. Autonomous AI & Semantic Search Expertise
- Google Cloud AI Ecosystem: Deep, practical experience with Vertex AI, Vertex AI Search, and the Google Agent Development Kit (ADK).
- Search & Retrieval: Advanced expertise in Vector Databases, Semantic Search engineering, and Google Grounding models.
- Agent Frameworks: Proven track record of designing autonomous, multi-turn, state-persisted AI agent workflows and multi-agent swarms.
3. Testing, Methodology & CI/CD Pipelines
- Deterministic & Agentic Testing: High proficiency with Python testing frameworks (pytest, asyncio test patterns) combined with advanced agent evaluation methodologies. Experience implementing Vertex AI Gen AI Evaluation Service or Google ADK Eval harnesses (managing Golden Datasets, tracking tool trajectory scores, response similarity, and tool_uses assertions).
- CI/CD Infrastructure: Hands-on experience building automated deployment gates using Google Cloud Build, GitHub Actions, Artifact Registry, and Cloud Run, ensuring that both deterministic microservices and stochastic AI prompts are versioned, regression-tested, and audited.
- Spec-Driven Development: Mastery of building software by first defining rigorous specifications, interfaces, and automated test schemas.
- AI-Assisted Mastery: High proficiency using advanced AI-assisted engineering workspaces (e.g., Cursor, GitHub Copilot) not as a crutch, but as a tool to maximize architectural execution speed.
This is a UK-based position only. To be considered for this role, you must already reside in the United Kingdom and hold full, unrestricted Right to Work in the UK.
The role operates on a Hybrid model based out of London. While we champion high-autonomy and flexible remote working patterns, regular in-person collaboration in London is a core requirement for our founding team dynamics. Applications that do not meet these geographic and right-to-work criteria will not be reviewed.
Pay: Up to £65,000.00 per year
Work Location: Hybrid remote in London
