MLOps and LLM Engineer, Agent Orchestration, Memory, Tracing (100% Remote) (Pakistan Only)

About AllShore Talent

AllShore Talent is a leading remote staffing company, offering top-tier professionals working 100% remote to businesses worldwide. Specializing in IT and software development, design, administrative support, digital marketing, and more. AllShore connects organizations with skilled talent to meet diverse business needs.

About The Client
The Client is a full-service digital solutions partner committed to building websites and apps that are as unique as the businesses they represent. With a deep belief in the value of strong client relationships, their approach is grounded in trust, collaboration, and a genuine understanding of each client’s goals. From custom web and app design to comprehensive eCommerce strategies, branding, marketing, and social media management, they offer tailored solutions that help businesses grow and thrive online. Whether supporting startups, medical practices, or real estate firms, they reject cookie-cutter templates in favor of scalable, intuitive, and beautifully designed platforms. Their team works closely with each client, providing hands-on support and flexible service options that make managing an online presence easy, impactful, and aligned with business objectives.

About The Role
We are looking for an MLOps/LLM Engineer to lead the orchestration of AI agents, manage vector-based memory systems, and implement robust monitoring and tracing for model performance. Your work will directly power the AI features our clients use, ensuring reliability, scalability, and adaptability.

Project context
You will operationalize LLM features for a RAG and agent based platform, with vector memory in Pinecone and graph style knowledge retention.

What you will do

• Build agent orchestration flows with LangChain or comparable frameworks

• Implement graph memory and long term context for agents, including vector and relational stores

• Set up evaluation and tracing for prompts, agents, and RAG quality, LangSmith, OpenTelemetry, custom dashboards

• Manage embeddings pipelines and index lifecycle, ingestion, chunking, versioning

• Productionize models with reproducible environments, Docker and CI CD

• Control cost and latency with caching, batching, and routing across providers

• Partner with backend to expose robust APIs, and with QA to define LLM specific tests

What you will bring

• 4 plus years in ML engineering or MLOps, strong Python

• Proven work with LLM providers and embeddings, OpenAI, Anthropic, HF, or Vertex AI

• Experience operating vector databases and data pipelines

• Familiarity with evaluation frameworks for RAG and agents

• Cloud experience on GCP, AWS, or Azure

Nice to have

• Knowledge graphs or graph databases, Neo4j or similar

• Policy and guardrail tooling, content safety, PII redaction