Pierre-Louis Bescond – Medium

Pierre-Louis Bescond

Pinned

Pierre-Louis Bescond
in
Towards Data Science

Explaining how I reached the top ranks of the new Data-Centric competition

Following the endorsement of Andrew NG himself(!) regarding my last article, it felt natural to share all tips (with code!) of how I…

9 min readSep 5, 2021

--

5

Explaining how I reached the top ranks of the new Data-Centric competition

--

5

Pinned

Pierre-Louis Bescond

Pierre-Louis Bescond’s articles on Medium

Data Science, Machine Learning and Innovation

6 min readJan 2, 2021

--

Pierre-Louis Bescond’s articles on Medium

--

Pierre-Louis Bescond
in
Roquette Data & Adv. Analytics

Retro-Engineering a Database Schema: Mistral Models vs. GPT4, LLama2, and Bard (Episode 3)

How do “Mixtral 8x7b” and “Mistral Au Large”, these new LLMs developed by former developers of LLama-2, perform compared to OpenAI, Google…

11 min readMar 5, 2024

--

Retro-Engineering a Database Schema: Mistral Models vs. GPT4, LLama2, and Bard (Episode 3)

--

Pierre-Louis Bescond
in
Towards Data Science

Benchmarking Snowflake Cortex against Scikit-Learn on a real-life forecasting use-case.

One of the most trending cloud-based Data platforms, Snowflake, now embeds advanced modeling features and I gave a shot to the forecasting…

9 min readFeb 25, 2024

--

Benchmarking Snowflake Cortex against Scikit-Learn on a real-life forecasting use-case.

--

Pierre-Louis Bescond
in
Roquette Data & Adv. Analytics

My first impressions on Copilot for Microsoft 365

Here is what the first few hours of working with this long-awaited add-on have taught me…

9 min readJan 25, 2024

--

My first impressions on Copilot for Microsoft 365

--

Pierre-Louis Bescond
in
Towards Data Science

Retro-Engineering a Database Schema: GPT vs. Bard vs. LLama2 (Episode 2)

In my previous article, I benchmarked GPT-4 model against Bard. Now Llama-2 enters the arena and it’s high time we see how it performs…

6 min readOct 6, 2023

--

Retro-Engineering a Database Schema: GPT vs. Bard vs. LLama2 (Episode 2)

--

Pierre-Louis Bescond
in
Roquette Data & Adv. Analytics

The Long-Term Paradox of LLMs Training Material

Large Language Models have marked a turning point in content creation, but in the long term, how will we manage ever more AI-based corpus…

4 min readOct 6, 2023

--

The Long-Term Paradox of LLMs Training Material

--

Pierre-Louis Bescond
in
Towards Data Science

Retro-engineering a database schema and quality checks: GPT vs. Bard

Can LLMs retro-engineer a consolidated dataset to design the original database and suggest the corresponding data quality checks?

9 min readJul 25, 2023

--

Retro-engineering a database schema and quality checks: GPT vs. Bard

--

Pierre-Louis Bescond
in
Towards Data Science

Querying a Corpus of Documents in GPT Mode with Azure “Prompt Flow”

How to automatically vectorize content and create LangChain-like mechanisms to efficiently query a corpus of documents

6 min readJul 21, 2023

--

Querying a Corpus of Documents in GPT Mode with Azure “Prompt Flow”

--

Pierre-Louis Bescond
in
Roquette Data & Adv. Analytics

Data Quality on steroids with Google Dataplex and BigQuery combined

Google Cloud Platform keeps on releasing exciting features within its Data suite. Solving one of the top challenges of Data teams, ensuring…

6 min readMay 9, 2023

--

Data Quality on steroids with Google Dataplex and BigQuery combined

--

Pierre-Louis Bescond

Pierre-Louis Bescond

Head of Data & Advanced Analytics @ Roquette | Winner of the 1st WorldWide Data Centric Deep Learning Contest | Data Science & Machine Learning Passionate!

Following

Help
Status
About
Careers
Blog
Privacy
Terms
Text to speech
Teams