AI Engineer

Company Description BtSavvy builds FORGE, a control layer for AI-enabled engineering that embeds architecture, security, and governance into the path to production. The company focuses on ensuring that AI-generated software is reliable, compliant, and traceable from requirements through implementation, especially for complex, large-scale systems. FORGE allows architecture, security, and platform teams to define standards once, then applies them consistently across projects while developers continue to use their preferred AI tools. BtSavvy works inside clients’ own repositories, cloud accounts, and controls, so code, data, and applications remain fully owned and managed by the client. The company also offers modular services such as AI Strategy & Architecture, Build, Data Foundation, Cloud, DevSecOps, Governed Enablement, Run, and Assurance, often starting with a FORGE Diagnostic to deliver a working slice of a workload in weeks.
Role Description Every role here works on one problem: making software built with AI safe to ship and supportable in production. In this role you design and build agentic systems that run inside enterprise operations. You hold them to the standards a regulated business expects: reliable, observable, secure and affordable to run. You work alongside client teams, teach the discipline as you go, and put your name on what reaches production.
What you'll do
Craft and delivery
- Lead the design and build of AI and ML systems end to end, from the first architecture decision to production at scale.
- Build agentic systems that behave predictably, with a clear scope, checks outside the model and a record of what they did.
- Build and evolve the platforms, pipelines and infrastructure that make model development and deployment repeatable across teams.
- Set the standard for AI-assisted engineering with tools such as Claude Code and Cursor, and help the team use them well.
Collaboration and communication
- Work inside client teams as a forward-deployed engineer, in their repositories, cloud accounts and controls.
- Partner with product, engineering and leadership to shape AI strategy and tie technical direction to business outcomes.
- Explain AI trade-offs, risks and opportunities in plain language to technical and non-technical people.
- Lead design reviews and technical discussions that raise the bar for engineering rigour.
Leadership and influence
- Define AI architecture and engineering standards, including responsible AI practice.
- Mentor junior and mid-level engineers through coaching, code review and pairing on hard problems.
- Take on the most ambiguous, highest-stakes work and drive it to production with care for reliability, cost and safety.
- Teach client engineers how we work, so the capability stays with them after we leave.
What you'll bring
- 7+ years of professional software engineering, including 4+ years building AI/ML systems in production and deep hands-on work with generative AI.
- Expert software engineering in Python or a similar language, with a strong design sense for scalable, maintainable systems.
- Deep experience with at least one major cloud platform. AWS and its generative AI services are a strong advantage. We also deliver on Azure, Google Cloud and Oracle Cloud.
- A record of designing and shipping complex agentic systems in production.
- Command of AI frameworks and orchestration tools.
- Strong experience with evaluation frameworks and observability for LLM applications, including building them where they don't yet exist.
- A deep understanding of AI safety and responsible AI, including prompt-injection defenses, PII handling and limiting what an agent is allowed to do.
- Extensive experience building RAG pipelines: chunking, embedding models, vector databases and advanced retrieval.
- API design experience, including architecting and integrating internal and third-party services at scale.
- Cost optimization expertise: token economics, caching, model routing and quantization.
- Strong working knowledge of Docker and Kubernetes.
- Daily, expert use of AI coding tools such as Claude Code and Cursor.
How to apply
Send us a message on LinkedIn with your CV. Add a short note about one agentic system you took to production: what it did, what went wrong, and what you changed.

Work arrangement
Yes