Need a LLM AI developer

About the Role

We are looking for an LLM Developer who can work across both model tuning and data engineering/scraping pipelines. You will shape model behaviour, optimise GPT-based systems, and build scalable datasets that power our AI models.

This role is ideal for someone who enjoys building at the intersection of AI model optimisation + data pipelines.

Key Responsibilities
1. Model Tuning & Optimization

Tune GPT-family models through structured prompts, system instructions, and parameter adjustments (temperature, top-p, top-k, penalties, max tokens, stop sequences).

Experiment with fine-tuning approaches (LoRA/QLoRA, SFT, OpenAI fine-tuning).

Build reusable prompt templates and instruction frameworks.

Create automatic evaluation pipelines for consistency, factuality, and reliability.

2. Data Scraping & Dataset Creation

Build custom scrapers using Python (BeautifulSoup, Scrapy, Playwright, Selenium).

Collect structured and unstructured data from multiple sources.

Clean, preprocess, label, and convert scraped content into training datasets.

Automate recurring scraping tasks with robust error handling and logging.

3. Data & Model Pipelines

Develop retrieval systems using embeddings and vector databases (Pinecone, Weaviate, Qdrant, FAISS).

Implement chunking, ranking, and retrieval optimisation strategies.

Build internal tools to visualise model outputs and dataset quality.

4. Deployment & Integration

Integrate models into backend services via APIs.

Optimise inference performance, latency, and cost.

Monitor model behaviour over time and iterate based on feedback data.

5. Safety & Quality Systems

Implement guardrails for model responses.

Build evaluation scripts for hallucination detection, quality scoring, and reliability checks.

Skills & Requirements
Must-Have

Strong Python skills (Scrapy/BeautifulSoup/Playwright/Selenium).

Experience tuning GPT models (prompt tuning + parameter optimisation).

Understanding of HuggingFace, PyTorch, or OpenAI fine-tuning APIs.

Experience building datasets for LLM training or RAG.

Familiarity with embeddings and vector databases.

Ability to run fast experiments and debug model behaviour.

Good-to-Have

Experience with vLLM, LlamaIndex, LangChain.

Exposure to distributed training or optimisation frameworks.

Knowledge of cloud services (AWS/GCP), Docker, API design.

Who You Are

Analytical and curious about how LLMs behave.

Strong builder mindset — quick with prototypes, fast with iteration.

Comfortable working in ambiguity and solving end-to-end problems.

Self-driven with high ownership.

Why Join Us?

Build core AI capability for a next-generation product.

Work on challenging LLM tuning and large-scale data problems.

High-speed environment with direct impact and autonomy.

Opportunity to grow into a lead LLM/AI engineer role.