---
url: https://fin.ai/events/ai-series/building-frontier-ai-products
title: "Building frontier AI Fin x Cognition x Harvey x Perplexity"
description: "Builders and decision-makers share real lessons, challenges, and breakthroughs across the full AI stack."
---

In person & Virtual

YBCA Forum, San Francisco

Thursday Sept 11, 2025

# Building frontier AI productswith Fin x Cognition x Harvey x Perplexity

## Missed it live? Watch the recording to hear AI experts break down how leading AI products are built, from training specialized models to scaling real-world applications.

![Fin](https://cdn.sanity.io/images/4bgok82s/production/404add042d7a65d59c49dbff11b3850fd6ab0ee1-16x16.svg)

![Cognition](https://cdn.sanity.io/images/4bgok82s/production/dae565e36558dbcd361aca18fcf0e5c852c5e34c-63x49.svg)

![Harvey](https://cdn.sanity.io/images/4bgok82s/production/3d86a3fc740dda7c11086bbb14d15ba8aa072f4c-64x64.svg)

![Perplexity](https://cdn.sanity.io/images/4bgok82s/production/b0de9aa2708138c388b2915bc4abe3103e27b796-184x56.svg)

On demand

Featured Speakers

![Silas Alberti - Founding Team](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2F75ec22254b189ae6427460092a75ea53ff6aa38b-256x305.webp&w=512&q=90)

* * *

Silas Alberti

* * *

Founding Team,  
Cognition

![Niko Grupen - Head of Applied Research](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2F13cbd816dc9843c4b859c8af3c2f8127f09600d8-256x305.webp&w=512&q=90)

* * *

Niko Grupen

* * *

Head of Applied Research,  
Harvey

![Brett Chen - Tech Lead & Member of Technical Staff](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2Fc72da4cdca0483a0c710817efd41b9dfd9eadaee-256x305.webp&w=512&q=90)

* * *

Brett Chen

* * *

Tech Lead & Member of Technical Staff,  
Perplexity

![Fergal Reid - Chief AI Officer](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2F28d3615074c732f2aab59f9d88b93b8b90d364af-256x304.webp&w=512&q=90)

* * *

Fergal Reid

* * *

Chief AI Officer,  
Fin

![Molly Mahar - Principal AI Designer](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2Fd45838410911ee5d163cce1ebfcb62c07855c590-256x305.webp&w=512&q=90)

* * *

Molly Mahar

* * *

Principal AI Designer,  
Fin

![Jordan Neill - SVP Engineering](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2F604e8c2e0a487534e99cfef1bf077332c7ac8f36-256x305.webp&w=512&q=90)

* * *

Jordan Neill

* * *

SVP Engineering,  
Intercom

Explore the models launched at this event[Learn more](/cx-models)

AI Research

Go deeper with the AI team: building the models

![](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2F47879e0167b476e2d32aa74cc88ff331a9a06338-448x207.webp&w=512&q=90)

Finetuning Retrieval for Fin  
[Learn more](https://fin.ai/research/finetuning-retrieval-for-fin/)

At Intercom, we’ve built Fin, an AI-powered support bot designed to understand users’ issues and answer their questions accurately. To do this, Fin relies on state-of-the-art large language models (LLMs)...  
[Learn more](https://fin.ai/research/finetuning-retrieval-for-fin/)

![](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2F3a2b45cac4d39665894fc69c315bbdf6a90ce95a-448x207.webp&w=512&q=90)

Using LLMs as a Reranker for RAG: A Practical Guide  
[Learn more](https://fin.ai/research/using-llms-as-a-reranker-for-rag-a-practical-guide/)

How to use parallel LLM-based reranker in RAG: improve AI agent answers, reduce latency, and see the exact prompt we deployed.  
[Learn more](https://fin.ai/research/using-llms-as-a-reranker-for-rag-a-practical-guide/)

![](/_next/image?url=https%3A%2F%2Fcdn.sanity.io%2Fimages%2F4bgok82s%2Fproduction%2Fc1259f7fafef93416dd6ae49c8892b1c61797830-448x207.webp&w=512&q=90)

How We Built a World-Class Reranker for Fin  
[Learn more](https://fin.ai/research/how-we-built-a-world-class-reranker-for-fin/)

Inside Fin AI agent for customer support: Fine-tuning ModernBERT reranker for RAG, outperforming Cohere v3.5, while cutting latency and costs  
[Learn more](https://fin.ai/research/how-we-built-a-world-class-reranker-for-fin/)
