AI Engineer

Located onsite in DHA Phase 3, Lahore, this role operates from 1 PM to 10 PM and requires 2 to 4 years of relevant experience. The position involves contributing to an innovative AI-driven no-code platform that transforms ideas, prompts, and designs into fully functional web and mobile applications. As an AI Engineer, you will be responsible for designing and delivering the core intelligence that interprets user needs and generates user interfaces, backend logic, and databases. Additionally, you will research and develop alternatives to leading model APIs, including self-hosted and on-premise large language models (LLMs).

Key Responsibilities

Engineering:
- Design and develop LLM-powered pipelines that translate natural language requirements into application code, data models, and APIs.
- Build multi-step AI agents capable of planning, code generation, self-correction, and automated bug fixing.
- Create design-to-code features supporting inputs from tools like Figma, Adobe XD, and wireframes using vision and multimodal models.
- Engineer prompt systems, context management, and retrieval-augmented generation (RAG) to ensure outputs adhere to the component library and coding standards.
- Develop evaluation frameworks to continuously measure and improve the quality, accuracy, and reliability of generated code.
- Optimize model latency, cost, and token usage across various providers such as OpenAI, Anthropic, and open-source alternatives.
- Collaborate closely with frontend, backend, and product teams to integrate AI capabilities into the platform’s builder experience.

Research & Exploration:
- Investigate and assess alternatives to frontier model APIs, including open-source LLMs like Llama, Qwen, Mistral, and DeepSeek for code generation and reasoning.
- Design and implement on-premise and private-cloud LLM deployments to address data privacy, cost management, and enterprise client needs.
- Benchmark self-hosted models against leading models focusing on accuracy, latency, and cost, providing recommendations for platform integration.
- Explore model optimization techniques such as quantization, distillation, and fine-tuning methods (LoRA/QLoRA) to enable efficient operation on limited hardware.
- Prototype hybrid routing systems that delegate simple tasks to smaller in-house models and complex tasks to frontier models.
- Stay updated with the latest AI research and translate promising concepts into prototypes and internal technical documentation.

Required Qualifications
- Proficiency in Python, Go, and/or TypeScript programming languages.
- Hands-on experience developing production applications using LLM APIs.
- Familiarity with agent frameworks such as LangChain, LangGraph, LlamaIndex, or similar, along with vector databases.
- Experience with abstract syntax tree (AST) parsing, code analysis, or compiler technologies.
- Practical knowledge of serving open-source LLMs using tools like vLLM or Ollama.
- Understanding of GPU infrastructure, model memory requirements, and inference scaling techniques.
- Experience working with structured outputs, function/tool calling, and automated code generation.
- Exposure to cloud deployment environments (AWS or Azure) and containerization using Docker.

Preferred Qualifications
- Experience in fine-tuning or evaluating open-source models.
- Previous involvement with no-code, low-code, or developer tool products.
- Knowledge of MCP or tool-integration protocols.

This role offers a unique opportunity to work at the intersection of AI research and software engineering, contributing to cutting-edge technology that empowers users to build applications without writing code. If you have a passion for AI, software development, and innovation, this position provides a dynamic and challenging environment to grow your expertise.