LLM Research

I am currently a Member of Technical Staff at Flourish Labs.

Previously I worked on LLM pretraining, post training, agentic search/retrieval, and evals at Databricks Research (2023-2026).

Some fun research projects at Databricks included:

• OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning
• KARL: Knowledge Agents via Reinforcement Learning
• LoRA Learns Less and Forgets Less (TMLR 2024, ICLR 2025) with Dan Biderman
• Beyond Chinchilla-Optimal: Accounting for Inference in Language Model Scaling Laws (ICML 2024) with Nikhil Sardana
• DBRX, an open weights 132B/36B MoE. See this fun Wired article about how we were open weights SOTA for ~3 weeks before Llama 3 was released!: Inside the Creation of the World’s Most Powerful Open Source AI Model

As a Research Scientist at MosaicML (2021 - 2023), I was part of the team that pretrained and finetuned the open-source large language models MPT-7B and MPT-30B at the dawn of the ChatGPT era.

Back when the MosaicML NLP team consisted of only 9 researchers, we did some work on optimizing BERT pretraining. Here is our detailed blog post and report: MosaicBERT: A Bidirectional Encoder Optimized for Fast Pretraining (NeurIPS 2023). We used a lot of the insights from this work to build MPT-7B and MPT-30B. This work formed the backbone of nomic-bert and ModernBERT.

This talk by Jonathan Frankle gives an overview of some of MosaicML’s early days.

Selected LLM Papers & Technical Blog Posts

• 2026 — OfficeQA Pro: An Enterprise Benchmark for End-to-End Grounded Reasoning
  Krista Opsahl-Ong, Arnav Singhvi, Jasmine Collins, Ivan Zhou, Cindy Wang, Ashutosh Baheti, Owen Oertell, Jacob Portes, Sam Havens, Erich Elsen, Michael Bendersky, Matei Zaharia, Xing Chen. (arXiv preprint).
• 2026 — KARL: Knowledge Agents via Reinforcement Learning
  Jonathan D. Chang, Andrew Drozdov, Shubham Toshniwal, Owen Oertell, Alexander Trott, Jacob Portes, Abhay Gupta, Pallavi Koppol, Ashutosh Baheti, Sean Kulinski, Ivan Zhou, Irene Dea, Krista Opsahl-Ong, Simon Favreau-Lessard, Sean Owen, Jose Javier Gonzalez Ortiz, Arnav Singhvi, Xabi Andrade, Cindy Wang, Kartik Sreenivasan, Sam Havens, Jialu Liu, Peyton DeNiro, Wen Sun, Michael Bendersky, Jonathan Frankle. (arXiv preprint).
• 2025 — Retrieval Capabilities of Large Language Models Scale with Pretraining FLOPs
  Jacob Portes, Connor Jennings, Erica Ji Yuen, Sasha Doubov, Michael Carbin (NeurIPS Workshop).
• 2025 - Improving Retrieval and RAG with Embedding Model Finetuning
  Jacob Portes, Andrew Drozdov, Erica Ji Yuen, Vincent Chen, Sean Kulinski, Milo Cress, Colton Peltier, Sam Havens, Michael Carbin, Vitaliy Chiley and Connor Jennings
• 2024 — Long Context RAG Performance of Large Language Models
  Quinn Leng*, Jacob Portes*, Sam Havens, Matei Zaharia, Michael Carbin (NeurIPS Workshop).
• 2024 — LoRA Learns Less and Forgets Less
  Dan Biderman, Jacob Portes, Jose Javier Gonzalez Ortiz, Mansheej Paul, Philip Greengard, Connor Jennings, Daniel King, Sam Havens, Vitaliy Chiley, Jonathan Frankle, Cody Blakeney, John P. Cunningham. (TMLR).
• 2024 — Introducing DBRX: A New State-of-the-Art Open LLM
  Mosaic Research Team.
• 2024 — Beyond Chinchilla-Optimal: Accounting for Inference in Language Model Scaling Laws
  Nikhil Sardana, Jacob Portes, Sasha Doubov, Jonathan Frankle (ICML).
• 2023 — LIMIT: Less Is More for Instruction Tuning Across Evaluation Paradigms
  Aditi Jha, Sam Havens, Jeremy Dohmann, Alex Trott, Jacob Portes. (NeurIPS Workshop).
• 2023 — MosaicBERT: A Bidirectional Encoder Optimized for Fast Pretraining
  Jacob Portes*, Alexander Trott*, Sam Havens, Daniel King, Abhinav Venigalla, Moin Nadeem, Nikhil Sardana, Daya Khudia, Jonathan Frankle (NeurIPS).
• 2023 — MPT-30B: Raising the Bar for Open-Source Foundation Models
  MosaicML NLP Team.
• 2023 — Introducing MPT-7B: A New Standard for Open-Source, Commercially Usable LLMs
  MosaicML NLP Team.
• 2022 — Fast Benchmarking of Accuracy vs. Training Time with Cyclic Learning Rates
  Jacob Portes, Davis Blalock, Cory Stephenson, Jonathan Frankle (NeurIPS Workshop).

---

Computational Neuroscience

Brain Machine Interfaces and Biological Learning Rules

During my PhD I worked on biologically plausible learning in recurrent neural networks (RNNs), reinforcement learning (RL), and motor control with James M. Murray: “Distinguishing Learning Rules with Brain Machine Interfaces” (NeurIPS 2022).

The Fly Connectome

For a large part of my PhD, I worked on a project with Rudy Behnia, Larry Abbott and Jessica Kohn on the neural computation of motion in Drosophila eyes. Our paper “Flexible filtering by neural inputs supports motion computation across states and stimuli” was published in Current Biology. Here is a Current Biology “Dispatch” that summarizes this work: Motion vision: Pinning down motion computation in an ever-changing circuit

Our work is summarized in this research talk:

Some of my pre-PhD work in the Hillman Lab investigated patterns of neural activation and blood flow (i.e. neurovascular coupling) in the rodent cortex.

In a previous life, I wrote a review-style master’s thesis on superconducting qubits for quantum computing and dabbled in philosophy of science.