Sr ML Engineer (Remote, Contract) [HR216] (PK)

Smart Working; Saddar; À distance; PKR 33 223 000 - 47 065 000; Plein temps

Résumé du poste

Smart Working is seeking a Senior ML Engineer to build and modernise an AI platform, focusing on productionising NLP and generative AI systems. You will deploy, monitor and operate ML models and pipelines in a remote-first, globally distributed team.

You will work on RAG-enabled capabilities, model evaluation and observability, collaborating with backend engineers to deliver scalable ML services. The role emphasizes production-grade ML architecture, security, governance and end-to-end ML

Qualifications

• 6+ years of professional AI/Machine Learning experience with production exposure.
• 5+ years of professional MLOps experience.
• At least 2+ years of real Applied AI experience, beyond experiments.
• Strong Python experience; core language for this role.
• Proven productionising and deploying AI/ML applications and models.
• Experience with generative AI/LLMs and evaluation for grounding/hallucination.
• Experience building ML pipelines and production architectures.
• Hands-on with goldens datasets and evaluation gating.
• Observability and telemetry for ML systems.

Responsabilités

• Refactor, modernise and productionise ML models and Applied AI capabilities including NLP and generative AI.
• Build new ML components and modular production-ready solutions.
• Develop production ML applications and supporting services using Python.
• Maintain ML pipelines for model integration, evaluation, deployment and operation.
• Design observable ML workflows with retry logic and error handling.
• Create model evaluation pipelines using goldens and quality thresholds.
• Evaluate models using metrics for generative AI, classification and other use cases.
• Implement guardrails to assess grounding and hallucinations in outputs.
• Apply Applied AI techniques including RAG where appropriate.
• Design provenance for models, prompts and data for auditability.

Connaissances

Python

MLOps

Applied AI/ML

NLP

Generative AI

RAG

Model evaluation

Production ML

Backend engineering

Cloud platforms

Docker

Kubernetes

FastAPI

Argo Workflows

Outils

Docker

Kubernetes

FastAPI

Argo Workflows

AWS

Azure

GCP

Description du poste

At Smart Working, we believe your job should not only look right on paper but also feel right every day. This isn’t just another remote opportunity — it’s about finding where you truly belong, no matter where you are. From day one, you’re welcomed into a genuine community that values your growth and well‑being.

Our mission is simple: to break down geographic barriers and connect skilled professionals with outstanding global teams and products for full‑time, long‑term roles. We help you discover meaningful work with teams that invest in your success, where you’re empowered to grow personally and professionally.

Join one of the highest‑rated workplaces on Glassdoor and experience what it means to thrive in a truly remote‑first world.

About the Role

We are seeking a Senior ML Engineer with strong experience in Applied AI, Machine Learning and MLOps to build and modernise an AI platform.

The role combines Applied AI, MLOps and backend/platform engineering, with a strong focus on productionising, deploying, evaluating and operating ML/AI systems. You will build new ML capabilities, modernise existing NLP and generative AI systems, and create reliable, observable infrastructure that makes models easier to integrate, evaluate, monitor and deploy.

Responsibilities

• Refactor, modernise and productionise existing ML models and Applied AI capabilities, including NLP and generative AI solutions.
• Build new ML components and re‑engineer existing models into standardised, production‑ready modular components.
• Develop production ML applications and supporting services primarily using Python.
• Build and maintain reliable ML pipelines covering model integration, evaluation, deployment and operation.
• Engineer resilient ML workflows with appropriate retry logic, error handling and repeatable execution.
• Design and automate model evaluation pipelines using golden datasets and appropriate quality and performance thresholds.
• Evaluate different types of models using metrics appropriate to their outputs, including generative AI, classification and other ML use cases.
• Implement appropriate guardrails and evaluation mechanisms to assess grounding, hallucinations and quality of generative AI outputs.
• Apply Applied AI techniques, including RAG, where appropriate to the ML capabilities being developed.
• Design mechanisms for model, prompt and input‑data provenance to support auditability and reproducibility.
• Build infrastructure supporting shadow testing, A/B testing, fallback strategies and kill switches for safe ML deployment.
• Support the labelling, curation and ongoing development of golden datasets used for model evaluation.
• Build structured human‑in‑the‑loop feedback pipelines to capture reviews and corrections and improve ML datasets.
• Integrate third‑party AI APIs and build appropriate adapter/API interfaces.
• Implement observability and telemetry covering model behaviour, errors, compute costs, token usage and latency.
• Contribute backend engineering capability required to integrate ML components reliably into the wider application.
• Support both batch and real‑time ML workloads as the platform develops.

Requirements

• 6+ years of professional AI/Machine Learning experience, with genuine production experience.
• 5+ years of professional MLOps experience.
• At least 2+ years of real Applied AI experience, working with AI/ML capabilities beyond experimentation or personal projects.
• Strong professional Python experience; Python is the core programming language for this role.
• Proven experience productionising and deploying AI/ML applications and models.
• Strong understanding of both Applied AI/ML and MLOps, rather than experience limited solely to model research or experimentation.
• Strong hands‑on experience with model evaluation and defining appropriate quality/performance criteria for production ML systems.
• Experience working with generative AI/LLMs and understanding evaluation considerations such as grounding and hallucination.
• Hands‑on understanding of RAG and other Applied AI techniques.
• Experience building and operating ML pipelines and production ML architectures.
• Experience designing reliable ML workflows with appropriate error handling, retry mechanisms and repeatable execution.
• Experience working with golden datasets and using them for model evaluation and quality gating.
• Experience building observable ML systems using appropriate logging, monitoring and telemetry.
• Understanding of model/data provenance, auditability and reproducibility.
• Experience implementing safe production deployment practices for ML systems, including appropriate testing, fallback or fail‑safe mechanisms.
• Sufficient backend engineering experience to build APIs, integrations and production‑ready services around ML capabilities.
• Experience solving real production ML problems, including reliability, deployment, integration, evaluation or performance challenges.
• Familiarity with governance, compliance and safeguards relating to sensitive data and AI‑generated outputs.

Nice to Have

• Experience with FastAPI for building Python‑based ML APIs.
• Exposure to Argo Workflows or similar DAG‑based orchestration frameworks.
• Experience with Docker and Kubernetes.
• Experience working with one or more major cloud platforms: AWS, Azure or GCP.
• Multi‑cloud or cloud‑agnostic application experience.
• Experience or understanding of TypeScript and/or Go.
• Production experience with speech‑to‑text or transcription models.
• Experience working with real‑time ML applications.
• Experience with traditional NLP models, transformer‑based models, encoders and decoders.
• Experience integrating external models/providers such as OpenAI or Claude.

We may use artificial intelligence (AI) tools to support parts of the hiring process, such as reviewing applications, analyzing resumes, or assessing responses and identifying potential inconsistencies or verification signals in application materials based on available information. These tools assist our recruitment team but do not replace human judgment. Final hiring decisions are ultimately made by humans. If you would like more information about how your data is processed, please contact us.