AI Architect

We are looking for a Senior Agentic AI Engineer who doesn’t just experiment with LLMs, but designs, builds, evaluates, and ships production-grade agentic systems.

This role is about owning the design and implementation of an Agentic AI engine that automates the software development lifecycle end to end: from requirements and design to coding, testing, and artifact quality.

This is not a “prompt tinkering” position. This is systems engineering, at scale, with real constraints and real users.

Your core responsibility is the hands-on design and implementation of a production-ready agentic AI engine that:

• Automates key stages of the SDLC
• Uses multi-agent architectures
• Is measurable, debuggable, and scalable
• Performs reliably in real-world scenarios, not just demos

Technical Requirements

We expect strong hands-on experience, not theoretical familiarity:

• Hands-on experience designing and developing agentic systems using LangGraph, LangSmith, LangChain, Pydantic, Braintrust.dev, Arize, OpenRouter, or similar frameworks
• Proven experience building production-grade agentic systems, including:
  • Agent engineering
  • Prompt engineering
  • Context engineering and context usage optimization
  • Designing and building offline and online evals
  • Designing and implementing agent tools (including MCP)
  • Memory engineering
• Experience rapidly iterating on agentic workflows:
  • Formulating hypotheses
  • Adjusting evals
  • Tuning system prompts
  • Building specialized agent tools to hit defined performance targets
• Experience designing REST APIs and integrating with external APIs
• Strong proficiency in Python and FastAPI
• Hands-on experience with Agentic AI-powered coding tools, such as:
  • GitHub Copilot, Cursor, Windsurf, Cline, RooCode, Claude Code, OpenAI Codex
  • Experience customizing agentic workflows, rules, and prompts is a strong plus
• Experience or strong understanding of:
  • Secure and safe agentic systems
  • Guardrails
  • LLM / agentic threat modeling
  • Hardening and red teaming

Responsibilities

• Design and implement a state-of-the-art agentic AI engine that automates the software development lifecycle
• Actively participate in defining and analyzing requirements for the agentic engine and the surrounding ecosystem
• Design and implement multi-agent squads across different stages and dimensions of the SDLC
• Apply best industry practices in agent engineering, including:
  • Task breakdown and execution
  • RAG
  • Memory management
  • Process and artifact quality evaluation
• Develop and maintain:
  • Architecture vision
  • Agentic architecture
  • Design and technical documentation
• Integrate the agentic AI engine with other systems and platform components
• Communicate clearly and effectively with stakeholders and external clients

You will be building core infrastructure, not features on the edge.

This role is for someone who wants to define how agentic systems are built, evaluated, and operated, and who understands that reliability, observability, and iteration speed matter as much as raw model capability.

In short:
you’ll be building agents that actually work, not just look impressive in a demo.