Principal AI Verification Engineer
Location
Cadiz / Spain
Type
Full time
Department
Engineering
Overview
Predictable Machines is seeking a Senior AI Verification Engineer to architect and lead the development of mathematically rigorous AI verification systems. We're building the next generation of trustworthy AI—where every output is traceable, verifiable, and backed by formal reasoning.
We're tackling a fundamentally different challenge: Rather than building faster or more capable AI, we're ensuring AI systems produce provably correct, auditable results. Our platform combines streaming verification architectures, formal methods (SMT solvers), and functional programming principles to create AI systems that enterprises can actually trust with critical decisions.
We need someone who:
Architects verification-first systems—designs streaming architectures, event-driven workflows, and real-time verification pipelines using Kotlin, TypeScript, and functional programming principles.
Integrates formal methods with AI workflows—combines SMT solvers, logical reasoning engines, and mathematical validation tools to create provably reliable AI systems.
Builds enterprise-grade verification platforms—develops Server-Sent Events APIs, MCP protocol implementations, TypeScript client libraries, and Docker-based tool ecosystems that other engineers actually want to use.
Leads mathematical AI reliability—implements multi-layered verification (factual, logical, mathematical) with complete audit trails, source provenance, and confidence scoring systems.
Thinks in systems and proofs—comfortable with both distributed systems architecture and mathematical reasoning; excited about making AI outputs as reliable as traditional software systems.
Predictable Machines values engineers who understand that the hardest problem in AI isn't making it work—it's making it trustworthy. We're building verification systems that could power the next generation of AI-driven critical infrastructure. If you're passionate about combining cutting-edge AI capabilities with mathematical rigor and systems engineering discipline, we want to meet you.
Key responsibilities
Architect verification-first AI systems with our CTO and engineering team—design streaming verification pipelines, formal reasoning workflows, and mathematically rigorous AI validation systems that meet enterprise reliability requirements.
Lead formal verification methodology development—integrate SMT solvers, logical consistency engines, and mathematical validation tools into cohesive verification architectures; establish standards for AI output correctness and auditability.
Drive technical innovation in AI reliability—research and implement novel approaches to real-time fact-checking, logical verification, and mathematical validation; contribute to advancing the state-of-the-art in trustworthy AI systems.
Build enterprise verification platforms—architect Server-Sent Events APIs, TypeScript client libraries, Docker-based tool ecosystems, and MCP protocol implementations that enable seamless integration of verification capabilities into customer applications.
Optimize streaming verification performance—design high-throughput event processing systems, efficient verification tool orchestration, and scalable containerized architectures that can handle enterprise-scale verification workloads with minimal latency.
Establish verification engineering practices—define code quality standards, testing methodologies, and deployment practices for systems where mathematical correctness and audit trails are mission-critical; mentor junior engineers in functional programming and formal methods approaches.
Qualifications
Required Experience:
3+ years building production AI verification or reliability systems—demonstrated experience architecting systems that validate, audit, or formally verify AI outputs rather than just integrating AI capabilities.
Strong functional programming background—proficiency in Kotlin, TypeScript, or Scala with understanding of immutable architectures, type-safe design, and compositional system building.
Enterprise systems architecture experience—hands-on work with streaming APIs, event-driven architectures, containerized deployments, and building systems that other engineers integrate with.
Mathematical validation or formal methods exposure—practical experience with constraint solvers, logical reasoning systems, mathematical verification tools, or a systematic approach to correctness validation.
Preferred Qualifications:
Advanced degree in Computer Science, Mathematics, or Logic—with coursework in formal methods, theorem proving, model checking, or mathematical verification approaches.
Experience with verification toolchains—SMT solvers (Z3, CVC4), theorem provers, constraint satisfaction systems, or mathematical validation frameworks.
Streaming systems expertise—Server-Sent Events, real-time event processing, WebSocket architectures, or high-throughput data pipeline design.
MCP protocol or AI tool integration—experience building tools that AI assistants can discover and invoke, or similar plugin/extension architectures.
Enterprise authentication and multi-tenancy—OAuth2, session management, audit logging, and building systems that meet enterprise security requirements.
Essential Mindset:
Systems-first thinking—approaches AI as a component in larger, reliable systems rather than the primary focus; understands that verification infrastructure is as important as AI capabilities.
Mathematical precision—comfortable with formal reasoning, logical consistency requirements, and building systems where correctness can be proven rather than just demonstrated.
Collaborative architecture—excels at designing APIs and interfaces that other engineers want to use; values code quality, documentation, and maintainable system design.
Enterprise empathy—understands business requirements for audit trails, compliance, and reliability; can translate mathematical verification concepts into business value.
Continuous learning mindset—stays current with formal methods research, verification tooling advances, and enterprise AI reliability practices while maintaining focus on practical implementation.
What we offer
Pioneer the future of trustworthy AI systems—work alongside world-class engineers and researchers building the mathematical foundations that will make AI reliable enough for critical infrastructure and enterprise decision-making.
Architect systems with lasting impact—your verification architectures, streaming protocols, and formal methods integrations will directly influence how enterprises can safely deploy AI at scale; work in a flat structure where technical decisions drive company direction.
Flexible, results-focused environment—remote-first culture that values shipping reliable verification systems over office presence; autonomy to design and implement complex technical solutions with minimal bureaucracy.
Equity in the verification revolution—performance-based compensation tied to building the infrastructure that could power the next generation of trustworthy AI; when enterprises adopt verified AI systems, everyone benefits.
Deep technical growth opportunities—dedicated time for formal methods research, mathematical verification exploration, and contributing to open-source verification tooling; access to cutting-edge research in AI safety and mathematical reasoning.
Continuous learning in an emerging field—professional development budget for conferences, courses, and research in AI verification, formal methods, and mathematical validation—areas where expertise is rare and highly valued.
Join us to build the verification systems that will make AI trustworthy enough to power critical business decisions, financial systems, and infrastructure. We're not just building better AI—we're building the mathematical foundations that ensure AI systems work correctly, transparently, and reliably when it matters most.