Verified Profile
AI Development Chatbot Development Research & Development Consultations Machine Learning Data Engineering

  • Summary

    Profile Status
    South Africa(South Africa Standard Time)
    Years in Business
    Sectors and Industries
    marketing real-estate ecommerce banking investment-banking

    About Us

    Data Engineering, AI Development, Chatbot Development, Data Analytics, Consultations. BI, Monitoring & Analytics, API Development, Machine Learning

    Years in Service
    Team Size
    Min contract size
    Avg hourly rate
  • What we do

    We are great at these things.

    LLM Finetuning NLP Messaging Embeddings
    AI Development
    We implement self-hosted, private models as well as integrations into full-service LLM hosts such as OpenAI.
    Chatbot Development
    Enhancing customer engagement and service through intelligent, conversational bots.
    Research & Development
    Optimizing LLM applications for utility using the latest releases and
    We help review AI strategies and share learnings from our experiences with stacking LLM features.
    Machine Learning
    training statistical models on data
    Data Engineering
    Building data pipelines
  • Our Previous Work

    Some of our previous work.

    Data Engineering

    Codebase Refactoring and Migration

    In a significant data migration project, Mooncoon was instrumental in transferring over 20TB of on-premises data to AWS S3 for a major South African bank. This migration centralized the bank's data on a robust, scalable cloud platform, enhancing its accessibility and security. Additionally, Mooncoon demonstrated proficiency in code translation and optimization by re-engineering actuarial SAS code to AWS Glue, enabling the actuarial team to leverage a more cost-effective, modern platform

    AI Chatbot Development

    Mooncoon developed a Retrieval-Augmented Generation (RAG) enabled chatbot for a biotech company. The chatbot leverages large language models (LLMs) like Llama2 and utilizes LangChain for model orchestration, Dagster for data orchestration, S3 for storage, and Qdrant as a vector database. The RAG architecture combines the benefits of a knowledge base and generative models to reduce hallucinations, maintain up-to-date information, and leverage domain-specific knowledge

    AI Development

    Real-time marketing AI system

    Mooncoon designed and implemented a real-time marketing AI system for a boutique agency. The system, built on Google Kubernetes Engine, utilizes ClickHouse, DBT, Airflow, and CatBoost to analyze Google and Facebook analytics, thereby informing strategic marketing decisions

  • How did we do?

    This is what our clients say about us.