Colin Lachance’s Post

As a super quick overview, this guide of tips and tricks is really useful to get you underway with AI - I highly recommend a quick read through.

Doubling the excitement! Legal analytics users are riding the generative AI wave with twice the enthusiasm. Dive into the data-driven buzz with Lex Machina's survey insights and request your own copy of the report today! https://bit.ly/433W4mX #LegalAnalytics #LexMachina #AI #LegalAI

Evaluate your fine-tuned LLMs. 🚀 A powerful approach to improving the applicability of LLMs to specific tasks is through fine-tuning. Once you've fine-tuned LLMs on specific data, it's essential to evaluate their performance and effectiveness in real-world scenarios. You can now easily fine-tune LLMs using the Clarifai Platform. After fine-tuning, you can also assess their performance using the newly introduced LLM Evaluation module. The module allows you to evaluate LLMs against standardized benchmarks and custom criteria. This guide covers a step-by-step process to fine-tune and evaluate your LLMs: Read more: https://lnkd.in/gDienp7w

Special treat this time (🐶) Dean Pleban from DagsHub will join us next week (29th August) to talk about how to evaluate LLMs. Agenda: - Main challenges in customizing and evaluating LLMs for specific domains and applications. - Review workflows and tools to help solve for those challenges. https://lnkd.in/dwnS_4Dw #genai #LLM #ai #LLMEvaluation

🚀 New Blog Alert! 🧠 I just published a blog on how to streamline LLM testing with a test-driven approach using Promptfoo. If you're looking to improve your prompt engineering, evaluate multiple LLMs, or ensure reliable outputs using local models, this is for you! I explore how to test LLMs efficiently using both custom-hosted and locally-run models, providing more flexibility and control over your AI workflows. Check it out and let me know what you think! 💡 #AI #LLM #MachineLearning #PromptEngineering #AItesting #TechBlog

How to convert vectors having different values into one hot encoding to use that data to train classification models. #AI #ML #Datascience #innovation

One of my most complete pieces, Document Intelligence for LLMs, explained with Claude 3.5 stack: https://lnkd.in/d_w_vsy5

Can GENAUDIT Enhance Fact-Checking for LLMs? HIGHLIGHTS GENAUDIT addresses LLM inaccuracies effectively. The tool provides document-based evidence validation. It maintains a balance between error detection and precision. #GENAUDIT #LLM Learn More: https://lnkd.in/dqy52XVh

LLMs (chatbots) can’t be trusted for financial advice - econ prof Gary Smith The LLM responses demonstrated that they do not have the common sense needed to recognize when their answers are obviously wrong https://lnkd.in/grfbkMDq It takes an experienced financial planner to distinguish between good and bad advice, so clients may as well skip the LLMs and go to the knowledgeable human.

Alongside the well-known RAGs, agents [1] are another popular family of LLM applications. What makes agents stand out is their ability to reason, plan, and act via accessible tools. When it comes to implementation, AdalFlow has simplified it down to a generator that can use tools, taking multiple steps (sequential or parallel) to complete a user query. https://lnkd.in/ez9fR7iY