Lead Full Stack Developer

Overview

Saatchi & Saatchi is seeking a Lead Developer for an innovative AI platform initiative.

You have 8+ years building production software systems and at least a few years leading or anchoring a team technically. You are a hands‑on builder and you hold the architectural picture of a multi‑tier platform — React frontend, Spring Boot backend, FastAPI agentic services, agent SDKs on top of LLM providers. You are equally comfortable reviewing a database migration, debugging a prod incident, mentoring a senior engineer, or sitting in a client architecture conversation. You thrive at the intersection of architecture, engineering, and program reality.

Responsibilities

• Own the technical direction across the platform's tiers — frontend, backend, agentic/AI services, data, and integrations — and hold the shared mental model of how the system actually fits together.
• Pair closely with the Senior Technical PM and Program Director to translate product intent into a credible technical plan, and to surface what is genuinely hard vs. what is just unfamiliar.
• Lead architecture and design reviews, write the deciding ADR when needed, and keep a small set of vendor‑neutral patterns the team can lean on instead of a sprawl of frameworks.
• Stay hands‑on in the code — pick up meaningful slices of work, especially across tier boundaries (e.g. backend agentic, frontend auth/RBAC, infra‑aware app changes).
• Be a first‑class code reviewer across the stack: you set the bar for clarity, testability, security, and operability, and you raise the team's bar by reviewing thoughtfully.
• Mentor mid‑level and senior engineers; coach them through technical decisions, on‑call experiences, and difficult code reviews.
• Partner with the Platform Engineer on production reliability — capacity, scaling, observability, and incident response — and own application‑side reliability concerns.
• Drive integration architecture with internal and external systems using API‑first and event‑driven patterns; ensure identity, authorization, data protection, and auditability are built in, not bolted on.
• Champion the LLM/agent layer — agent orchestration, tool/plugin design, prompt and evaluation loops, provider abstraction (e.g. Vertex AI, Claude on Vertex, OpenAI) — and keep us out of vendor lock‑in we cannot pay for.
• Build redundancy on yourself — pair, document, rotate ownership, and ensure no critical system has exactly one person who understands it.

Qualifications

• 8+ years in software engineering, with meaningful time as a hands‑on lead, staff engineer, or solutions architect on production systems.
• Proven experience designing and delivering complex, multi‑service platforms — not just one service or one tier.
• Strong backend depth in at least one modern stack (Java/Spring Boot, Python/FastAPI, Node, Go, .NET); able to ramp on the others.
• Solid frontend literacy — comfortable reading and reviewing modern React/TypeScript code and reasoning about state management, auth, and performance.
• Strong cloud architecture background, preferably GCP (we run primarily on GKE); willing to deepen wherever the platform takes us.
• Deep understanding of API design, integration patterns, distributed systems, and the failure modes that come with them.
• Experience with containerization, Kubernetes, and modern CI/CD; you do not need to be a DevOps specialist, but you must operate well in that world.
• Real production experience with AI / LLM systems — agents, RAG, evaluation, tool integration — or a strong, demonstrable plan for getting there.
• Comfortable working within Agile delivery models (Scrum, Kanban, or whatever variant actually ships).
• Excellent communication: you can explain a complex architecture to a non‑technical executive and disagree with a senior engineer respectfully and with evidence.

Required Soft Skills

• Hands‑on bias — you reach for the code, the logs, and the diagram before you reach for the meeting invite.
• Calm, decisive judgment when there is no obviously right answer.
• Strong written communication — you can write a one‑page architecture rationale that future‑you will still understand.
• Generous mentorship instinct — you make the people around you better, on purpose.
• Comfortable with ambiguity and willing to make reversible decisions quickly while flagging the irreversible ones for broader input.
• Pragmatic about technology choices — you optimize for the team's ability to operate the system, not for personal stack preferences.
• Honest about what you do and do not know; willing to say "I'll find out" and then actually find out.

Nice to Have

• Prior experience working inside Publicis Groupe or another large h