Cosine Blog
We write about our latest Model and Product developments, making software engineering more accesible and autonomous day by day.
AutoPM - The Autonomous Product Manager for Your Dev Backlog
Meet AutoPM: your AI-powered product manager that autonomously breaks down dev tasks and ships production-ready pull requests—no micromanagement needed.
Read the full article
Jun 12, 2025 — Alistair Pullen
AutoPM - The Autonomous Product Manager for Your Dev Backlog
Meet AutoPM: your AI-powered product manager that autonomously breaks down dev tasks and ships production-ready pull requests—no micromanagement needed.
Read the full article
Jun 12, 2025 — Alistair Pullen
All Posts
Cosine Invited to Help Shape the Future of UK AI Policy
Jun 9, 2025 — Yang Li
Newsroom
Making Prompting Faster with the New Genie Homepage
May 29, 2025 — Pandelis Zembashis
Changelog
Introducing Genie 2.1 - Faster, smarter and built for tougher tasks
May 22, 2025 — Pandelis Zembashis
Changelog
Mid May Genie Updates
May 13, 2025 — Pandelis Zembashis
Changelog
Genie 2.0 - The fastest, lightest and most capable autonomous software engineer.
May 6, 2025 — Alistair Pullen
Launch
Ghost in the Machine: From Mimicry to Mastery
Oct 31, 2024 — Giorgos Lysandrou
LLMs
Technical Report: Building Genie 1.0
Aug 12, 2024 — Alistair Pullen
Engineering
Cosine Genie 1 - SOTA AI Engineer
Aug 12, 2024 — Alistair Pullen
Launch
Lessons Learned from Integrating AI into Developer Tools
Jul 23, 2024 — Pandelis Zembashis
Product
A silent killer hidden amongst our indexer for months
Jun 21, 2024 — John Ingram
Engineering
Cosine Cloud
Jan 17, 2024 — Pandelis Zembashis
Launch
You (Kind of) Don't Need RAG - ChatGPT Finetuning Technique
Oct 17, 2023 — Alistair Pullen
LLMs
How Mythbusters helped us scale our RAG.
Apr 4, 2023 — Alistair Pullen
Engineering LLMs
I got early access to ChatGPT API and then pushed it to it’s limits. Here’s what you need to know.
Mar 1, 2023 — Alistair Pullen
LLMs
Why going viral forced us to rip out everything in a weekend
Feb 22, 2023 — Alistair Pullen
LLMs Product
Use GPT-3 "incorrectly" and reduce costs by40x while increasing performance by 5x!
Feb 6, 2023 — Alistair Pullen
LLMs
Three LLM tricks that boosted embeddings search accuracy by 37%
Dec 25, 2022 — Alistair Pullen
LLMs