Jianfeng Chi

Researcher
DISC Type : Cs

Research Scientist at Meta

New York, New York, United States

Overview

Jianfeng is a Research Scientist at Metas Superintelligence Labs, specializing in LLM post-training, reasoning, and safety alignment. He holds a Ph. D. in Computer Science from the University of Virginia and has contributed to key projects like Llama 3 and Llama Guard.

He has a deep interest in the practical application of AI, having previously worked on AI for production systems at Meta and as an intern at Amazon Web Services.

Jianfeng has published research on privacy attacks related to trajectory data in point-of-interest recommendation systems.

Personality Overview

Self-Disciplined

Cost Conscious

Process Focused

Being observant comes to them naturally.  The only way to convince them is by showing them examples and ample proof. They do not like taking risks at all and go for proven options in the end.

Topics They Care About

LLM Safety
His current work focuses on safety alignment, and he was part of the teams that developed Llama Guard 2 and Llama 3 to mitigate model safety issues.
AI Reasoning
He researches how to elicit 'aha moments' in LLMs to improve reasoning, particularly for safety alignment in RLHF.
Data Privacy
He has co-authored and shared a research paper on privacy attacks concerning trajectory data used in recommendation systems.

Media Appearances

Jianfeng has no verified media appearances

Work History

9-2022
Research Scientist at Meta
8-2017 - 8-2022
Graduate Research Assistant at University of Virginia
6-2021 - 8-2021
Software Engineering Intern, Machine Learning (PhD) at Facebook
5-2019 - 8-2019
Applied Scientist Intern at Amazon Web Services (AWS)

Education

2017 - 2022
Doctor of Philosophy (Ph.D.) from University of Virginia
2013 - 2017
Bachelor's degree from Beijing University of Posts and Telecommunications

More Information

Social Presence :

Prographics :

Exp : 5 Location : New York, New York, United States Job Level : N/A Designation : Research Scientist at Meta
URL has been copied!

Insights For Selling To Jianfeng

During A Call Or A Meeting

DO's

  • Actively address their concerns around change, risk, and acceptance by users
  • Share whitepapers or case studies that showcase measurable results instead of just telling customer stories
  • Share a one-off customer success story but keeps the focus on highlighting objective, numerical results

DONT's

  • Don't ask them to move fast, let them take their time and digest all the information
  • Avoid winging it with them particularly, answer a question only if you know the answer well
  • Avoid emotional and informal language, stay objective and to the point instead

When Cold Calling

When Writing An Email

While Negotiating & Closing

    The secret to closing fast with Jianfeng is

  • Adoption by others is very important to them, followed by confidence in ROI.
  • Will you ever get a clear answer from Jianfeng

  • They don’t say no often, they push out the decisions or keep going around in circles.

Insights For Deal Planning

    How fast (or slow) will Jianfeng move?

  • They don't like to hasten, so their speed of decision-making may be slow.
  • Can Jianfeng take some risk or not?

  • They have little willingness to take risks, and prefer making calculated decisions.

You And Jianfeng

Personality Compatibility


Other Meta Employees

Explore more public profiles from related professionals at the same organization.

More Profiles

Discover additional public profiles from our index.

Search more profiles

Looking for someone else? Search here for anyone.

Or visit Humantic AI to know more.