Our client is building an internal LLM platform for multiple business units (Arabic/English). They need an AI Specialist / ML Engineer to productionize models, stand up retrieval, and harden safety/observability—working closely with Security, Data, and Product.
Role at a glance
Own the end-to-end technical path from prototype to production for LLM features (chat, agents, document QA), with strong focus on reliability, safety, and cost.
Key responsibilities
Stand up inference endpoints and routing across OSS/commercial models; implement prompt/version management.
Build retrieval (chunking, embeddings, vector DB), caching, and eval pipelines; define SLOs for latency, cost, and quality.
Implement guardrails (PII/redaction, toxicity, jailbreak resistance) and red-team tests; instrument hallucination & grounding metrics.
Create golden-path IaC/Terraform templates; integrate with CI/CD, secrets/KMS, and observability (tracing, tokens, costs).
Optimize GPU/accelerator usage; benchmark models; support fine-tuning/LoRA where justified.
Partner with Security/Compliance on PDPL/data-residency and audit readiness.
Candidate profile
5–8+ years in ML/Platform with 2–3+ years LLM apps in production.
Strong Python + one typed language (TypeScript/Go/Java), vector DBs, and cloud (AWS/Azure/GCP).
Experience with eval/guardrail frameworks and observability; comfort with infra basics (containers, K8s).
Clear communication; thrives in cross-functional delivery.