Sonam Pankaj
where else to find them
what they talk about
all 6 in the filters
where they keep going back
- PyData Global3 editions2020–2024
- Berlin Buzzwords2 editions2024–2025
all 6 in the filters
on a stage with
who talks about the same things
- Akshay Ballal2 talks
- Mary Grygleski95 talks
- Zain Hasan6 talks
- Raphael De Lio8 talks
- JP Hwang4 talks
- Atita Arora8 talks
| title | conference | date |
|---|---|---|
| User Signal Dies at the Retrieval Boundary | AI Engineer World's Fair 2026 | 2026-06-29 |
Claims a concrete fix: utility-ranked memory that re-ranks retrieval by outcome history, not just semantic similarity, raising helpful memories and lowering… | ||
| Text Search on Images with Quantized ColPali | Berlin Buzzwords 2025 | 2025-06-15 |
ColPali is a late‑interaction vision‑LLM pipeline fine‑tuned on Pali Gemma that embeds whole page images so you can skip OCR, chunking and parsing; relevance… | ||
| Vector Streaming: The Memory Efficient Indexing for Vector Databases | PyCon DE & PyData 2025 | 2025-04-23 |
Rust-based Embed-Anything implements vector streaming by moving embedding work into a separate thread/process and communicating via MPSC channels, keeping only… | ||
| The Memory Efficient Indexing for Vector Databases | PyData Global 2024 | 2024-12-03 |
Claims a working vector streaming pipeline that parallelizes parsing, semantic chunking, embedding generation and upserts so embeddings never accumulate in… | ||
| The Unsung Hero of Vector Database: Metric Learning | Berlin Buzzwords 2024 | 2024-06-09 |
Metric learning is presented as the fix behind better vector search: using Siamese and triplet architectures with contrastive and triplet loss to stop… | ||
| Let chatGPT decide and run the function! | PyData Global 2023 | 2023-12-06 |
Claim: chatGPT can pick the right tool and return call-ready arguments via openai function calling. Demo uses GPT-3.5 (GPT‑4 noted as supported), registers… | ||
| Biases in Language Models | PyCon DE & PyData Berlin 2022 | 2022-04-11 |
Claims measurement of gender bias across Flair BERT and GloVe using Winograd-schema sentences about occupations and applies projection and fine-tuning to… | ||
| TimeSeries Forecasting with ML Algorithms and there comparisons | PyData Global 2020 | 2020-11-11 |
Compares concrete algorithms on a Covid19 kaggle univariate confirmed‑cases set using RMSE and shows ARIMA giving the lowest error RMSE 4000 while moving… | ||
| Time Series Forecasting on CoronaVirus Dataset | PyCon India 2020 | 2020-10-02 |
Live work uses the coronavirus confirmed-cases time series and compares naive forecasting, moving average, linear regression, SVM, LSTM, ARIMA and Facebook… | ||
| nothing matches | ||