AI Architect – Software

Job Title: AI Architect – Software & Data Engineering, LLM/RAG Specialist

Location: Hybrid (3 days onsite / 2 days remote -Karachi or Islamabad)

Type: Full‑time

About the Role

We are looking for a hands-on AI Architect who bridges software development, data engineering, and advanced AI/ML. You will design and code AI solutions that leverage large language models, embeddings and retrieval‑augmented generation while ensuring robust data pipelines and integration with our software products. This role requires a strong foundation in software engineering and data engineering along with deep expertise in AI/ML.

Key Responsibilities

1. Cross‑Disciplinary Solution Design

○ Engage with stakeholders to understand product requirements and architect

end‑to‑end solutions incorporating data ingestion, model training, deployment

and monitoring.

○ Design systems that combine traditional software components with AI/ML models

and modern data engineering practices.

1. Hands‑On Development

○ Write and maintain production‑quality code across back‑end services, data

pipelines and model inference layers, primarily using Python and at least one

other language (e.g. .Net, Java).

○ Build RESTful or gRPC APIs to expose AI functionality and integrate with

front‑end and other microservices.

1. Data Engineering & Pipeline Construction

○ Develop and manage data pipelines for ingestion, transformation and storage

(using tools such as Apache Spark, Kafka, Airflow, SQL/NoSQL databases).

○ Work with vector databases and embedding stores for retrieval‑augmented

systems.

○ Ensure data quality, schema management and robust ETL processes that

support model training and inference.

1. Advanced AI/ML & LLM Expertise

○ Train, fine‑tune and deploy machine‑learning and deep‑learning models,

including large language models (LLMs) and transformer architectures (e.g.,

GPT, Llama, Mistral).

○ Implement embedding techniques (e.g., sentence and document embeddings),

retrieval‑augmented generation (RAG) pipelines and hybrid search (vector +

keyword).

○ Build customised pipelines using frameworks like LangChain, LLamaIndex or

similar to integrate models with diverse data sources.

1. System Optimisation & Scalability

○ Optimise model performance (latency, throughput and accuracy) for production

workloads, using techniques such as model quantisation, batching and caching.

○ Scale services using containerisation (Docker) and orchestration (Kubernetes) on

cloud platforms (AWS, Azure, GCP).

1. Collaboration & Mentorship

○ Collaborate with software engineers, data scientists and DevOps teams to deliver

cohesive solutions.

○ Mentor junior engineers and share best practices in software development, data

engineering and AI/ML.

1. Monitoring & Maintenance

○ Implement observability for data pipelines and AI services (logging, metrics,

model drift detection).

○ Continuously monitor and iterate models and pipelines based on feedback and

new data.

Required Skills & Qualifications

● Bachelor's or Master's degree in Computer Science, Data Engineering, Artificial

Intelligence or related field.

● Experience: Minimum 6–8 years in software development and data engineering, with at

least 3 years of hands‑on AI/ML experience.

● Proficiency in Python and deep familiarity with AI/ML frameworks (e.g., PyTorch,

TensorFlow, Scikit‑learn) and LLM libraries (e.g., Hugging Face Transformers,

LangChain).

● Hands‑on experience building and deploying large language models, embeddings, RAG

pipelines and vector‑database‑powered applications.

● Strong knowledge of data engineering tools (Spark, Kafka, Airflow), relational and

NoSQL databases, and proficiency in SQL.

● Experience designing and operating microservices, APIs and containerised applications

on cloud platforms.

● Solid understanding of MLOps best practices (CI/CD for ML, model versioning,

automated testing, monitoring).

● Strong problem‑solving and communication skills; ability to translate business

requirements into technical designs.

Preferred Qualifications

● Experience with GPU orchestration (NVIDIA Triton, KServe) for model serving.

● Knowledge of privacy‑preserving techniques and responsible AI practices.

● Exposure to reinforcement learning or recommendation systems.

What We Offer

● The chance to work on cutting‑edge AI projects integrating software, data and advanced

models.

● A collaborative environment where hands‑on coding and architectural leadership go

hand in hand.

● Support for ongoing learning and professional development in AI/ML and data

engineering.

● Health Insurance

● Gadget Allowance

Job Type: Full-time

Pay: Rs300, Rs350,000.00 per month

Work Location: In person