AI Engineer

We're looking for an AI Engineer to design, build, and ship agentic AI systems end-to-end - from early prototypes to production deployments.

You'll work closely with the founder and early collaborators, owning meaningful parts of system architecture, agent design, evaluation, and delivery.

This is a builder role, not research-only and not prompt-only.

What You'll Do

Build agentic AI systems

• Design and implement AI agents that reason, plan, and act across tools, APIs, and data sources
• Build multi-step workflows (ReAct-style, tool-calling, retrieval-augmented agents)
• Integrate LLMs into real operational systems (not just chat UIs)

Own systems in production

• Ship AI systems that are observable, debuggable, and reliable
• Design evaluation frameworks to measure agent quality, failure modes, and regressions
• Balance autonomy, control, cost, latency, and risk in real deployments

Applied AI engineering

• Design retrieval pipelines (RAG), embeddings, vector search, and hybrid approaches
• Improve system performance through prompt iteration, tooling design, and architectural changes
• Implement guardrails, fallback logic, and human-in-the-loop flows where appropriate

Collaborate & think in systems

• Translate vague business problems into concrete technical architectures
• Work across product, data, and domain constraints - not in isolation
• Contribute to internal patterns, templates, and reusable ClearCortex IP

Do you have experience in TypeScript?, Do you have a Master's degree?, Core skills

• Strong software engineering fundamentals
• Proficiency in Python and/or TypeScript
• Experience building and shipping production software

AI / agent experience

• Hands-on experience with modern LLM APIs (OpenAI, Anthropic, Gemini, etc.)
• Experience with agent frameworks (LangChain, LangGraph, LlamaIndex, or custom systems)
• Practical experience with:
• RAG and retrieval pipelines
• Tool use and function calling
• Multi-step reasoning workflows
• Prompt engineering beyond single prompts

Systems thinking

• Comfortable reasoning about:
• Failure modes
• Observability and logging
• Cost vs performance tradeoffs
• Autonomy vs control
• Ability to debug agent behaviour systematically (not by guesswork)

Nice-to-Have (Not Required)

• Experience with vector databases (Qdrant, Pinecone, Weaviate, etc.)
• Familiarity with evaluation techniques for LLMs and agents
• Experience deploying AI systems in cloud environments (Docker, CI/CD, basic DevOps)
• Prior work in applied ML, NLP, or decision systems
• Startup, studio, or early-stage product experience

You'll Thrive Here If You

• Prefer shipping systems over polishing slides
• Enjoy ambiguity and figuring things out from first principles
• Think in architectures, not just prompts
• Care about real-world impact more than benchmarks
• Want meaningful ownership, not ticket-driven work
• Are curious, pragmatic, and opinionated (but evidence-driven)

Why ClearCortex

• High ownership: You'll help shape how systems are built, not just execute tasks
• Real problems: Production systems with real users and real constraints
• AI-native environment: Agents, workflows, evaluation, and tooling are core - not bolt-ons
• Small, sharp team: Work directly with the founder, move fast, iterate constantly
• Long-term thinking: Reusable IP, platforms, and products - not throwaway projects

ClearCortex is an AI studio building production-grade AI systems for real businesses - not demos, not slideware. We design and deploy agentic systems, retrieval pipelines, and AI-powered workflows that operate in messy, real-world environments: incomplete data, unclear requirements, evolving goals, and human decision-makers in the loop. Our work sits at the intersection of: * AI agents & LLM systems * Applied software engineering * Real operational workflows (property, logistics, professional services, internal tooling) We're building reusable IP, internal platforms, and long-term products - while shipping high-impact client systems along the way. If you enjoy turning ambiguity into working systems, this role is for you.