AI / LLM Engineer

oladoc is seeking three passionate AI / LLM Engineers to join their team in Lahore, working full-time on-site at NASTP. This role is ideal for engineers who thrive in building and shipping real AI products rather than just experimenting with models. You will contribute directly to the AI stack behind Pakistan’s leading health platform and its new wellness brand, Fit by oladoc. The position demands a builder’s mindset focused on delivering production-ready solutions that impact millions of users and thousands of healthcare professionals.

Key Responsibilities
- Develop and scale LLM-powered features such as medical Q&A, symptom triage, document parsing, and clinical note summarization.
- Build and maintain Retrieval-Augmented Generation (RAG) pipelines over structured and unstructured healthcare data.
- Design and implement production-grade prompt engineering systems with versioning, testing, and performance monitoring.
- Create evaluation frameworks to measure model accuracy, reliability, latency, and output quality.
- Develop agentic workflows using tools like LangChain, LlamaIndex, and the Anthropic SDK.
- Integrate LLM capabilities into existing product workflows via APIs.
- Build internal AI tools to support operations, sales, and clinical teams.
- Ensure scalable AI services with appropriate guardrails, observability, and cost optimization.
- Take full ownership of your work from development through deployment and continuous improvement.

Required Qualifications
- Bachelor’s degree in Computer Science, Artificial Intelligence, Software Engineering, Data Science, or related fields, or equivalent practical experience.
- Strong Python programming skills with the ability to write clean, maintainable, production-ready code.
- Hands-on experience with LLM tooling such as OpenAI API, Anthropic Claude API, Hugging Face, LangChain, or LlamaIndex.
- Experience building APIs, integrating external services, handling JSON, debugging, and managing application environments.
- Solid understanding of embeddings, vector databases, retrieval pipelines, and LLM evaluation techniques.
- Familiarity with Git, testing, deployment workflows, and cloud infrastructure.
- Demonstrated ability to independently research, experiment, and solve problems.
- Proven builder mentality: identifying problems, developing solutions, shipping products, and owning outcomes.
- Production experience is highly valued; demonstrable projects and technical accomplishments matter more than formal qualifications alone.
- Must be available to work 100% on-site in Lahore, NASTP, with full-time commitment.
- Hands-on experience with LLMs is mandatory; candidates must be able to showcase real projects they have built.

Preferred Qualifications and Benefits
- Experience working in an AI-first engineering culture with tools like Claude, LangChain, and RAG pipelines already in production.
- Access to a full AI tool stack including Claude, Anthropic, and OpenAI API credits.
- Competitive compensation based on experience and technical capability.
- Free access to doctor consultations via oladoc’s network, lab discounts for employees and their families, and confidential mental health support through Shifa Therapy.
- Wellness benefits including nutrition coaching, fitness tracking, and wellness check-ins through Fit by oladoc.
- Direct mentorship from senior engineers and the CTO, providing real production exposure on a platform serving millions of users.
- Employee benefits such as health and life insurance, subject to company policies and eligibility.

This role is for engineers who treat AI as a working environment, not a weekend curiosity. Candidates must have built and deployed real LLM applications, understand prompt design, token budgets, latency, inference costs, and reliability in production. Familiarity with OpenAI, Anthropic, or Gemini APIs and the ability to navigate their documentation independently is essential. You should be able to evaluate model outputs critically and defend your technical decisions regarding RAG, fine-tuning, and other approaches.

Applicants are required to submit a CV along with either a GitHub link demonstrating their LLM work or a brief technical write-up describing an interesting LLM project, the problem it solved, and the technical decisions involved. Demonstrating how you have used LLMs to improve or stress-test your applications is a plus. Only candidates who provide proof of building and shipping real solutions will be considered.