Senior Applied AI Engineer, Forward Deployed

About the role

We are looking for a senior engineer to own our AI agent systems and deliver end‑to‑end capabilities from customer discovery through production operation. The role blends applied AI engineering, backend development, and direct engagement with business users on a WhatsApp‑first platform.

Key responsibilities

• Design and evolve agent orchestration, tool interfaces, conversational state, memory, retrieval and human hand‑off.
• Develop production Python services and LangGraph workflows, evaluate architectural trade‑offs and improve existing components.
• Lead technical discovery with merchants, translate business workflows into requirements and delivery plans, and own implementation, integration, rollout and post‑deployment verification.
• Engineer reliable handling of concurrent requests, retries, timeouts and external‑service failures while enforcing tenant isolation and secure data access.
• Establish evaluation datasets, regression suites and release criteria, diagnose failures and measure task completion, latency and operating cost.
• Provide technical leadership through design discussions, code reviews, documentation and knowledge sharing.

Required profile

• Proven use of AI‑assisted engineering tools to deliver tested software and ability to explain, debug and maintain the work.
• Senior‑level software engineering experience designing, delivering and operating customer‑facing systems.
• Hands‑on experience deploying AI or LLM applications to production with measurable operational results.
• Strong communication, independent technical judgment and ability to translate ambiguous requirements into concrete solutions.

Required skills

• Python, FastAPI, LangGraph
• PostgreSQL, Redis
• Google Cloud Platform (Cloud Run, Cloud SQL, Pub/Sub, Vertex AI)
• Docker, CI/CD, secrets management, monitoring
• WhatsApp Business Platform, Meta Cloud API, Twilio integrations
• Asynchronous processing, background jobs, event‑driven architecture
• LLM evaluation, retrieval, fine‑tuning, PyTorch, TensorFlow