AI Engineer
Company Description
LioByte Technologies builds AI systems and ships them into production for business clients.
Role Description
We need an expert in recommendation systems to design and ship our recommendation engine. You will own the work from the first data review to a live system that real users see. This is a 3 to 6 month consulting contract. If the work and the fit are strong, we will offer you a full time role.
You will use Claude Code every day to build faster, and you will review everything it writes.
Skills
Recommender Systems, Machine Learning, Python, Learning to Rank, Vector Search, SQL, Claude Code
What you will do
- Review our data and pick the right recommendation approach
- Design and build candidate generation and ranking
- Build a simple baseline first, then improve on it with proof
- Set up offline metrics and live A/B tests so we know what works
- Handle the cold start problem for new users and new items
- Serve recommendations through fast, reliable APIs
- Document your work and teach our team how to run and extend it
Top 5 technical skills we look for
- Recommendation algorithms: collaborative filtering, content based methods and hybrid approaches
- Ranking and learning to rank
- Embeddings and vector search
- Python and machine learning libraries (PyTorch, TensorFlow or scikit-learn)
- SQL, data pipelines and experiment design (A/B testing)
Must have
- 3+ years of experience in machine learning or software engineering
- Proven experience shipping a recommendation or personalization system to production
- Hands on experience with Claude Code
- Strong Python and SQL
- Clear thinking about metrics, such as precision, recall, NDCG and click through rate
- Able to work on your own and explain your decisions in plain language
Nice to have
- Experience with search systems such as Elasticsearch or Solr
- Experience with large language models for recommendations or reranking
- Experience with real time features and caching
How to apply
Apply on LinkedIn, or email your resume to hello@liobyte.com. Include a short note on the best recommendation system you shipped. Tell us what you built, what it moved and what you would do differently.