Full Stack AI/ML Engineer

Full Stack AI/ML Engineer - Karachi Onsite

Client Company Introduction:

Ireland based Leading Software Engineering Company delivering innovative, AI-powered platforms for global clients.

Job Description:

• Ship end-to-end features across front-end (React/Next.js) and back-end services (Node.js/Express or Python/FastAPI).

• Design and maintain relational database schemas (Postgres/MySQL), ensuring schema consistency, referential integrity, and performance.

• Develop systems that validate and optimize automatically generated code, ensuring it runs correctly against applications and databases.

• Integrate LLM capabilities into production applications, including embeddings, retrieval pipelines, and orchestration frameworks, with strong evaluation and guardrails.

• Own quality and reliability: testing (unit, integration, contract), observability (logs, metrics, traces), and performance optimization.

• Manage AI performance and cost trade-offs, ensuring scalability, accuracy, and compliance.

• Collaborate with product and engineering teams to deliver secure, scalable, and innovative solutions through CI/CD workflows.

Requirements:

• Hands on AI/LLM experience with proven production implementations (not just prototypes):

• -LLM APIs (OpenAI, Anthropic, or open-source).
  -Embeddings, RAG pipelines, and vector databases (pgvector, Pinecone, Weaviate).
  -Workflow/orchestration frameworks (LangChain, AutoGen, or equivalent).

• Strong full-stack engineering experience:

• Front-end: React / Next.js.

• Back-end: Node.js / Express and/or Python / FastAPI.

• Databases: Postgres/MySQL (schema design, constraints, indexing, migrations, optimization).

• Proven ability to debug, validate, and productionise AI-assisted workflows.

• Strong grounding in software engineering best practices (CI/CD, testing, version control, observability).

• Comfortable working in a fast-paced, innovation-driven environment.

Location: Shahra e Faisal, Karachi

Work Timings: Mon - Fri 11AM - 7:30PM

Work Mode: Onsite