Add profile generation use case #508
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
lightaime
reviewed
Apr 30, 2025
Comment on lines
61
to
83
| "user": ModelFactory.create( | ||
| model_platform=ModelPlatformType.OPENAI_COMPATIBLE_MODEL, | ||
| model_type="gpt-4.1", | ||
| api_key=os.getenv("OPENAI_API_KEY"), | ||
| model_config_dict={"temperature": 0.4}, | ||
| ), | ||
| "assistant": ModelFactory.create( | ||
| model_platform=ModelPlatformType.OPENAI_COMPATIBLE_MODEL, | ||
| model_type="gpt-4.1", | ||
| api_key=os.getenv("OPENAI_API_KEY"), | ||
| model_config_dict={"temperature": 0.4}, | ||
| ), | ||
| "content_researcher": ModelFactory.create( | ||
| model_platform=ModelPlatformType.OPENAI_COMPATIBLE_MODEL, | ||
| model_type="gpt-4.1", | ||
| api_key=os.getenv("OPENAI_API_KEY"), | ||
| model_config_dict={"temperature": 0.2}, | ||
| ), | ||
| "planning": ModelFactory.create( | ||
| model_platform=ModelPlatformType.OPENAI_COMPATIBLE_MODEL, | ||
| model_type="gpt-4.1", | ||
| api_key=os.getenv("OPENAI_API_KEY"), | ||
| model_config_dict={"temperature": 0.3}, |
There was a problem hiding this comment.
Can we use openai model platform instead?
lightaime
reviewed
Apr 30, 2025
Comment on lines
198
to
233
| html_prompt = """ | ||
| Generate a professional and natural-looking personal profile HTML page based | ||
| on the provided information, following these detailed requirements: | ||
|
|
||
| Use standard HTML5 structure. | ||
|
|
||
| Layout should be clearly organized using <h1> for the main title, <h2> for | ||
| section headings, and <p> for paragraph text. | ||
|
|
||
| Use proper <a href> hyperlinks for any external links (e.g., LinkedIn, | ||
| ResearchGate, Google Scholar). | ||
|
|
||
| Write in a natural, flowing narrative style suitable for introducing the | ||
| person to a broad audience (no bullet points). | ||
|
|
||
| Strictly follow the given content structure: | ||
|
|
||
| Personal Information (Name, Positions, Contact( include social media), | ||
| Research Keywords) | ||
|
|
||
| Summary (Short description of expertise and overall role) | ||
|
|
||
| Biography (Detailed career history (can be found in Linkedin) and research | ||
| focus) | ||
|
|
||
| Research Interests (Areas of active research) | ||
|
|
||
| Awards and Distinctions (Honors and recognitions) | ||
|
|
||
| Education (Academic degrees in order) | ||
|
|
||
| Engage / Related Links (Professional profiles and external resources) | ||
|
|
||
| The final HTML should look clean, modern, and ready to be used as a complete | ||
| professional personal page. | ||
| """ |
There was a problem hiding this comment.
Can we use response_format in ChatAgent to generate a pydantic object with required fields and use exact the same html template as https://cemse.kaust.edu.sa/profiles/ahmed-eltawil to render the information? Please do not use prompt to generate html.
lightaime
reviewed
Apr 30, 2025
Comment on lines
213
to
236
| html_user_prompt = f""" | ||
| Using the following JSON, build a complete HTML5 profile page. | ||
|
|
||
| • Sections: Personal Information, Biography, Research Interests, Awards | ||
| and Distinctions, Education. | ||
| Generate a professional and natural-looking personal profile HTML page | ||
| based | ||
| on the provided information, following these detailed requirements: | ||
| Using standard HTML5 + simple inline CSS, beautiful but not complicated. | ||
| Layout should be clearly organized using <h1> for the main title, <h2> for | ||
| section headings, and <p> for paragraph text. | ||
| Use proper <a href> hyperlinks for any external links (e.g., LinkedIn, | ||
| ResearchGate, Google Scholar). | ||
| Write in a natural, flowing narrative style suitable for introducing the | ||
| person to a broad audience. | ||
|
|
||
| Please write ALL the input content to the webpage. | ||
| JSON: | ||
| {profile.model_dump_json(indent=2)} | ||
| """ | ||
|
|
||
| html_response = html_agent.step(html_user_prompt) | ||
| html_code = html_response.msgs[0].content | ||
| html_code_block = re.search(r"```html(.*?)```", html_code, re.DOTALL) |
There was a problem hiding this comment.
Can we not use agent for generating HTML? Just use a pre-defined template for this
lightaime
reviewed
May 1, 2025
There was a problem hiding this comment.
Is this the updated generated example?
lightaime
reviewed
May 1, 2025
Comment on lines
243
to
252
| agent = ChatAgent( | ||
| system_message="""Reformat awards as HTML list items using the | ||
| structure, Like: | ||
| <li class="field__item"><strong>The | ||
| NSF CAREER grant in low-power computing and | ||
| communication systems</strong>, 2024</li>.""") | ||
| for award in awards: | ||
| response = agent.step(f"Reformat this: {award}") | ||
| formatted_awards.append(response.msgs[0].content.strip()) | ||
| return "\n".join(formatted_awards) |
There was a problem hiding this comment.
Do not do it in multiple steps. Use response format to format the awards in one shot. Otherwise, it is too exspensive
There was a problem hiding this comment.
The format problem can be solve by adding examples in the Field description.
lightaime
reviewed
May 1, 2025
Comment on lines
263
to
267
| items = [] | ||
| for degree in degrees: | ||
| response = agent.step(f"Reformat this: {degree}") | ||
| items.append(response.msgs[0].content.strip()) | ||
| return "\n".join(items) |
There was a problem hiding this comment.
Do not do it in multiple steps. Use response format to format the awards in one shot. Otherwise, it is too exspensive
There was a problem hiding this comment.
The format problem can be solve by adding examples in the Field description.
lightaime
reviewed
May 1, 2025
Comment on lines
183
to
192
| class AwardsAndDistinctions(BaseModel): | ||
| honors: list[str] = Field( | ||
| description="List of awards, honors, and professional distinctions " | ||
| "received") | ||
|
|
||
|
|
||
| class Education(BaseModel): | ||
| degrees: list[str] = Field( | ||
| description="Academic degrees in reverse chronological order, " | ||
| "including institution and year") |
There was a problem hiding this comment.
The format problem can be solve by adding examples in the Field description.
lightaime
reviewed
May 1, 2025
Comment on lines
269
to
277
| def format_links(links: list[str], html_template: str) -> str: | ||
| agent = ChatAgent( | ||
| system_message="Reformat link into this html structure: %s ." % | ||
| html_template) | ||
| items = [] | ||
| for link in links: | ||
| response = agent.step(f"Reformat this: {link}") | ||
| items.append(response.msgs[0].content.strip()) | ||
| return "\n".join(items) |
There was a problem hiding this comment.
Do not do it in multiple steps. Use response format to format the awards in one shot. Otherwise, it is too exspensive
nitpicker55555
force-pushed
the
profile-generation
branch
from
9700b6e to
df1ee1c
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Search a person's social media, Google Scholar, and other information to automatically build a profile webpage.