Applied AI Consulting

About

We are a group of experienced AI Engineers who have built AI products at leading tech companies. We specialize in building & improving LLM products. See our current and past clients here.

Problems We Solve

You should consider hiring us if …

  • You don’t know how to improve your LLM products systematically.
  • Your LLMs need to be cheaper or faster.
  • You are overwhelmed by tools & frameworks.
  • Feel lost on what to prioritize and what experiments you should run.
  • Need outside expertise to evaluate your needs and vet potential talent.

Services

We offer two tiers of services to support your goals.

Tier 1: Strategy

We will advise you on the following topics:

  • LLM performance issues: (cost, quality, speed)
  • RAG, Fine-Tuning, and Prompt Engineering
  • Upskilling your team: We will rapidly upskill and unblock your team with focused debugging sessions.
  • Strategy & Hiring: We will introduce you to talent, vendors, and partners in our network.
  • Evaluation Systems

We will give you ongoing feedback and guidance via regular meetings with your team. This will help you avoid common pitfalls and select the right tools and techniques, saving you time and money.

Tier 2: Comprehensive

Everything in Tier 1, plus:

  • Implementation: We will write production-ready code and/or prototypes to accelerate your AI product development.
  • Domain-Specific Evaluation Systems: We will design and implement custom evaluation systems to measure the performance and reliability of your LLMs.
  • Hands-On Model Optimization: We will fine-tune, prompt engineer and debug models to improve performance and efficiency.
  • Development Tools and Infrastructure: Build custom tools and infrastructure to streamline your AI development process.
  • Content and Writing: Produce written documents and blogs to communicate best practices, methodologies, and practical case studies to your stakeholders
  • Team Growth & Hiring: Work with 2-4 people individually on your team to rapidly upskill them on AI. We will also help you source and evaluate key hires.

Pricing

We work on a monthly retainer basis. The cost depends on the tier of service you choose:

  • Strategy: $183,600 for a 3-month engagement.
  • Comprehensive: $975,000 total for a 3-month engagement.

All of our work has a money-back guarantee1. Contact us at [email protected] to discuss starting an engagement. You can also book a paid call if you need immediate short-form advice.

Current & Past Clients

Members of our team have worked with the following companies:

  • Limitless AI: Limitless AI is a personal memory assistant that helps you remember, organize, and navigate your life.
  • Raycast: Raycast is a blazingly fast, totally extendable launcher. It lets you complete tasks, calculate, share common links, and much more.
  • Tensorlake: Build Knowledge for LLMs from un-structured data
  • Replicate: development of fine-tuning serverless infrastructure using axolotl and optimized LLM inference.
  • Weights & Biases: Provide product guidance for evaluation, annotation, and observability.
  • Honeycomb: Currently improving the natural language query assistant through evaluation systems and fine-tuning.
  • LangChain/LangSmith: provided product guidance for enterprise LLM tools.
  • Modal: Serverless tools for fine-tuning.
  • Rechat: We are working on Lucy, a conversational AI for real estate agents. You can read about my recent work on Rechat here.
  • Answer.ai: Conduct research on new LLM applications and help with product strategy.
  • Axolotl: Hamel is a core contributor to axolotl, a library for efficient fine-tuning of LLMs.
  • Kay.ai: Retrieve relevant context from the semantic web for your LLM apps with fully hosted embeddings.
  • Modal Labs: Modal specializes in cloud functions, offering a platform for running generative AI models, large-scale batch jobs, and more.
  • Pydantic: Pydantic provides data validation and settings management using Python type annotations, enforcing type hints at runtime with user-friendly error handling.
  • Posit: Helped Posit expand into the Python AI and ML ecosystem.
  • Catena: LLM infrastructure such as routing, evaluation, and orchestration.

Footnotes

  1. The quality of our work is guaranteed. If you do not believe we have met mutually established objectives, we will continue to work toward those goals with you for no additional fee. If, after such an additional attempt, you still believe we have not met your objectives, we will refund your fees in total.↩︎