AI Engineer

Job Description

About the Role 

We are hiring an elite cohort of highly hands-on AI Engineers to embed directly within Valsoft's Edelweiss Software Group portfolio of vertical market B2B SaaS companies. 

We are looking for builders — engineers who operate as system orchestrators rather than traditional syntax typists. You will design, implement, and deploy secure, enterprise-grade AI systems that automate workflows, eliminate manual processes, rapidly modernize massive legacy codebases, and create measurable business impact. 

You will work directly with operators, product teams, and leadership to turn inefficiencies into intelligent systems. Operating with a distinct startup mindset within a large-scale corporate portfolio, your primary focus is driving rapid product innovation, executing AI-driven legacy modernization via agentic swarm coding, and maximizing ROI through autonomous multi-agent workflows. 

If you enjoy owning problems end-to-end, possess a relentless bias toward action, and thrive on shipping real systems into production at an unprecedented velocity, this role is for you. 

Your mission: act as an organizational force multiplier to increase efficiency, reduce CAC payback periods, increase Net Revenue Retention (NRR), and unlock massive scalability. 

What You’ll Do 

Build & Ship AI Systems 

• Design, develop, and deploy production-grade, secure AI systems utilizing scalable polyglot microservices. 

• Integrate state-of-the-art enterprise LLMs (Anthropic Claude Opus 4.6, OpenAI GPT-5.2, Google Gemini 3) into SaaS platforms via dynamic, fault-tolerant API routing gateways. 

• Modernize extensive codebases using agentic swarm coding, automated refactoring tools, and cross-language translation. 

• Develop autonomous multi-agent workflows utilizing parallel orchestration tools like the OpenAI Codex and Claude Code CLI. 

• Automate comprehensive technical documentation for legacy and newly generated codebases using AI orchestration. 

• Rapidly prototype → validate via adversarial testing → deploy → iterate. 

Architect AI Infrastructure 

• Apply good knowledge of software engineering design patterns and architecture to ensure the scalability and maintainability of AI-generated systems. 

• Implement complex RAG systems and continually optimize trade-offs between massive-context hydration (up to 1-million tokens) and multi-stage semantic retrieval. 

• Work extensively with vector databases (Pinecone, Weaviate, FAISS) and manage persistent document embedding queues. 

• Build polyglot microservices specifically capable of handling massive, long-running operational tasks and streaming continuous token generation via persistent connection protocols (WebSockets and Server-Sent Events). 

• Ensure SOC2 compliance, data residency (via inference_geo parameters), and deterministic execution through policy-as-code agentic governance. 

• Deploy and scale models strictly within secure managed cloud boundaries (Azure AI, AWS Bedrock, GCP Vertex) leveraging Docker and Kubernetes orchestration. 

Transform the Application Lifecycle & Collaborate 

• Architect "zero-touch deployment" CI/CD pipelines fortified with adversarial gating. 

• Design and implement unit test and benchmark automation workflows, utilizing AI agentic coding to rigorously validate and test produced code. 

• Natively integrate synthetic red teaming into CI/CD to actively prevent prompt drift, reward hacking, and logic degradation. 

• Work directly with non-technical stakeholders to translate ambiguous business problems into secure, scalable AI solutions. 

• Optimize Atlassian workflows (Jira) to automate the translation of unstructured product requirements into structured prompt contexts for AI agents. 

Required Technical Skills 

Core Engineering 

• 3-5+ years of demonstrable enterprise software development experience. 

• Good knowledge of software engineering design patterns and architecture. 

• Flexible in most popular programming languages (Python, Javascript, Typescript, .Net, C#) and web frameworks (NextJS, etc.). 

• Ability to switch environments and programming using agentic programming tools to validate and test your produced code. 

• Backend development experience designing polyglot APIs, decoupled asynchronous microservices, and utilizing persistent connection protocols (WebSockets/SSE). 

• Familiarity with containerization (Docker, Kubernetes) and architecting multi-region, fault-tolerant cloud infrastructure (AWS, Azure, or GCP). 

AI & ML Systems 

• Enterprise LLM integration and dynamic API routing (OpenAI, Anthropic, Gemini, DeepSeek). 

• Multi-agent orchestration and advanced CLI tooling (Claude Code with Auto Hot-Reload and Forking Context, OpenAI Codex). 

• Mastery of localized AI IDE layers (Cursor for repository-wide reasoning, GitHub Copilot). 

• Massive-context RAG pipelines, dynamic chunking strategies, and vector databases (Pinecone, Weaviate, FAISS). 

• Custom evaluations (LangSmith), structured outputs, and managed fine-tuning in secure cloud boundaries. 

MLOps & Process Automation 

• Zero-touch CI/CD pipelines and advanced Git workflows (including Git worktree isolation for autonomous sub-agents). 

• Unit test and benchmark automation frameworks fully integrated with AI-driven testing. 

• Adversarial LLM testing and automated synthetic red-teaming integration. 

• Monitoring dynamic access limits and real-time atomic token/credit consumption tracking. 

• Strict hallucination mitigation, enterprise guardrails, and deterministic policy-as-code execution. 

Profile We’re Looking For 

• Highly hands-on — you ship, not just design. You operate at an unprecedented velocity by orchestrating advanced autonomous tools. 

• Entrepreneurial startup mindset — you see legacy inefficiencies and act with a relentless bias toward rapid action. 

• Comfortable with ambiguity and translating complex business needs into precise technical execution. 

• Business-oriented — you care deeply about ROI, operational leverage, and driving specific financial metrics (CAC, NRR). 

• Fast learner — keeps up with real-time shifts in AI ecosystems, hardware architectures (e.g., Cerebras chips), and vital API migrations. 

• Pragmatic — uses the right secure enterprise tool for the specific bottleneck, eschewing the fanciest public tool if it compromises stability or compliance. 

#EdelweissSoftware