Role-play AI agent

Posted on by Leave a comment

AI agent is amazing, promising, and useful in many business scenaria. With the increasing capability of large language model such as ChatGPT-4 and opensource models such as Llama3, it beccomes easy to develop LLM based agent to help the human to do tedious work. For example, it can replace coacher with AI agent to train the junior salesman, while the latter can play with the AI agent to improve their pitch skill. What you do is to select suitable opensource tools and models and to write promts, i.e. prompt engineering, for each application scenario. The following image is the screenshot of the role-agent demo what I build.

AI agent is salesman

Human user is a buyer

Ai agent is buyer

Human user is salesman

What you need to build the demo:

  • Streamlit: UI design
  • Ollama: manage open-source LLM.
  • LangChain: build LLM based chat pipeline
  • Prompt engineering: according to the task and role, write prompt.

Collection of prompt engineering

Posted on by Leave a comment

Prompty engineering is a critical part in developing AI agent. To make AI function as a particular role, e.g. insuarance salesman, insurance buyer, math teach, ….., it is necessary to write the task-specifc promt.

[Github] collection of prompt examples to be used with the ChatGPT model.

[Github] a curated list of the best system prompts for OpenAI’s ChatGPT, enabling developers and users to customize their AI’s behavior and interaction style

Resource collection of finance LLM

Posted on by Leave a comment

Finance knowledge graph

[Github] Financial Knowledge Graphs: small scale of financial knowledge graph, download stock data, visualize graph

[Github] Dynamic Knowledge Graph with Large Language Models for Global Finance

Large language model

[Github] AI4Finance Foundation, code, demo.

[Huggingface] FinGPT Forecaster, stock analysis give company ticker name

[Github] CFGPT: Chinese Financial Assistant with Large Language Model

Paper

A Survey of Large Language Models in Finance (FinLLMs)

BloombergGPT: A Large Language Model for Finance

MLOps roadmap 2024

Posted on by Leave a comment

The original content is in MLOps roadmap 2024, a wonderful article. I summarize the key points in the post for my reference.

IDDescriptionResources
1Programming
– Python & IDE
– Bash & command line editors		– IDE – visual studio code
– VIM
2Containerization and Kubernetes– Docker
– Kubernetes
3Machine learning fundamentalsa lot
4Machine learning fundamentalsBook: Introducing MLOps 𝖻𝗒 Mark Treveil 𝖺𝗇𝖽 Dataiku
5MLOps components– Git : Version control & CI/CD pipeline
– Airflow: Orchestration
– Mage-ai
– Mlflow: Experiment tracking and model registries
– Feast: feature store
– KubernetesPod Operator: Model training & serving
– nannyml: Monitoring & observability
– Evidently: Monitoring & observability
6Infrastructure as code– Terraform: Infrastructure as code

sympy: a python package for symbolic computation

Posted on by Leave a comment

Do you still remember how to calculate difference, gradient, limit, … in mathematic? if you forget, never mind. Here is a python package to help you. What you need: know python programming and define your functions, and then call package function to do your task. Symbolic computation is important component in PyTorch and Tensorflow.

The package name is symoy, refer Python package for Symbolic Computation. Symbolic computation deals with the computation of mathematical objects symbolically. This means that the mathematical objects are represented exactly, not approximately, and mathematical expressions with unevaluated variables are left in symbolic form. The following example

from sympy import *

x = symbols('x')

y = diff(sin(x)*exp(x), x)


y is 𝑒𝑥sin(𝑥)+𝑒𝑥cos(𝑥)

It is quite interesting. Will integrate into projects when symbolic processing is needed.

« Older Entries