Senior Platform Engineer, Data & AI Infrastructure

McKinney•Durham, NC

10h•$140,000 - $160,000•Hybrid

About The Position

We’re looking for a backend-leaning, Senior, Full Stack Engineer who will build AI-powered platforms, tools, and workflows that create value for our clients and empower our creative, strategy, operations, and account teams. You’ll design and build backend services, data-centric components, and internal tools, with a strong focus on Python and modern cloud infrastructure. You will be hands-on with integrating large language models (LLMs) and other AI capabilities into real products, from early design through deployment, monitoring, and iteration.

Requirements

Strong experience building backend services and APIs in Python (any modern web framework)
Experience with document databases (e.g., Firestore, MongoDB).
Containers & CI/CD: Docker/OCI image authoring, multi‑stage builds, image scanning/SBOMs, Artifact Registry; automated builds and deployments.
Cloud: GCP first (Cloud Run and Compute Engine; Secret Manager, Artifact Registry, Cloud Build/Deploy, Monitoring/Logging); Kubernetes familiarity welcome; equivalent AWS/Azure experience acceptable.
AI/LLM: Agentic architectures (tool/function use, multi‑step orchestration, retrieval/RAG, planners, memory), evaluation/guardrails/safety; experience with OpenAI, Anthropic, Google Gemini, and open‑weight models; familiarity with enterprise AI platforms that unify access to multiple model types.
APIs & Services: REST/GraphQL, schema/versioning, authentication/authorization.
Reliability: Testing (Pytest or similar), observability, performance tuning.
Frontend: Able to handle simple UI needs using modern web technologies; framework agnostic.
Process: Git‑based workflows and agile practices.
Communicates and collaborates effectively with creative, operations, strategy, and data partners.
Outcome‑oriented problem solving; balances speed, quality, and security.
Ownership and accountability; follows through and documents decisions.
Growth mindset; receptive to feedback and continuous learning.
Uses AI assistants responsibly with validation: evaluates outputs critically, adds tests, and adapts code to team conventions before submission.
4+ years of professional software engineering with a backend focus.
Proven and demonstrable experience building Python (FastAPI/Starlette) services and APIs for cloud deployment (GCP preferred).
Hands-on SQL experience in BigQuery; document database experience; Dataform exposure is a plus.
Prior experience integrating LLMs in an agentic manner into production apps or adjacent ML systems.

Nice To Haves

Kubernetes familiarity is a plus.
Dataform exposure is a plus.

Responsibilities

Design, build and maintain backend services and APIs primarily in Python (FastAPI/Starlette), emphasizing clean design, performance, and reliability.
Model data and write high‑quality SQL (primarily in BigQuery); use document databases (e.g., Firestore, MongoDB) where appropriate.
Build, harden, and operate containerized services: author Dockerfiles (multi‑stage), manage image versions in Artifact Registry, and enforce container security/scanning.
Deploy on GCP with Cloud Run and Compute Engine; leverage Secret Manager, Artifact Registry, Cloud Build/Deploy, and Cloud Monitoring/Logging; Kubernetes familiarity is a plus.
Integrate LLM/AI capabilities with an agentic approach (tool/function calling, multi‑step orchestration/planning, retrieval/RAG, and memory) using providers such as OpenAI, Anthropic, and Google Gemini, as well as open‑weight models; implement evaluation, safety, and guardrails.
Utilize our enterprise AI platform (Abacus.ai) that provides unified access to multiple language, image, and short‑form video models, plus prompt/version management, safety, and analytics; help define reusable patterns and abstractions for it across products.
Collaborate with data partners on ELT pipelines; use BigQuery and Dataform for transformations and analytics use cases.
Define and version API contracts (REST/GraphQL); document systems and interfaces.
Apply security and privacy best practices (authn/z, IAM least‑privilege, secret handling, input validation, rate limiting).
Establish observability (metrics, logs, traces) and conduct performance tuning; participate in pragmatic on‑call as needed.
Write tests (unit/integration/e2e); maintain CI/CD pipelines; conduct code reviews; mentor junior engineers