Jaime López Feo

CEO at The Agile Monkeys

[The Agile Monkeys @theagilemonkeys]

AI Monkeys Sessions #1 — What we learned getting certified by Anthropic.
A technical AI after-work session by The Agile Monkeys (Anthropic partner) covering real-world use cases, Agents Architecture foundations, lessons learned, plus Q&A and networking. Limited seats 🧵:

[Javi™ @javitm]

@rauchg I literally shipped a CLI today 😅🎉

https://t.co/xbw6ezGRLi

[Javi™ @javitm]

This week I ran a research agent to get a SOTA idea on the best of the best for confidential audio transcriptions on macOS.

It found a bunch of very nice tools, libraries and models, but I thought… why not build another small composable one I can use as an agent tool?

The agent one-shotted this one, written in Swift, using optimized models for the Apple Neural Engine, distributed via brew. So I guess this is our new reality…

[Jaime López Feo @jlo2412]

@ohryansbelt Awesome @ohryansbelt We (@theagilemonkeys) built PrivateGPT (appears in your image), although we had to change the name...
So many memories... Thanks a lot for sharing this. We loved it.

[Javi™ @javitm]

Here are the first two real use cases where we successfully applied Autonomous Researcher:

1. Optimizing an e-commerce search engine: the agent searched for the latest information retrieval research papers, generated evaluation datasets and experiments, and tested them with real code. After a month, it achieved ~30% improvement on eval metrics.

2. Extracting article ideas from live Slack conversations: We connected it to a few internal channels. The agent accumulates information in its knowledge base, compares it against state-of-the-art public info, and returns drafts for review. We publish or reject them with reasons; rejections feed back into the knowledge base, improving the odds of interesting ideas next time.

[Javi™ @javitm]

We’ve been experimenting with research agents for a while now and decided to open source Autonomous Researcher — a framework built for long-running agents that accumulate memory, taste, and traceable progress over weeks.

Persistent Knowledge Base with validated Hypothesis → Experiment→ Findings loops, Challenge/Strategic reviews, and durable artifacts that survive interruptions.

Not just faster hill-climbing → deeper, reproducible research.

Works with Codex + Claude Code.

Starter template here → https://t.co/TP6qkQP5Hi

#AIagents #AutonomousResearch #AgenticAI

[Javi™ @javitm]

Love this. We’ve been exploring a related direction with Autonomous Researcher — basically taking the “human edits prompt / agent edits code / eval decides” loop and extending it into persistent research memory.

Instead of just a better https://t.co/NXmbbozNTP, the agent has to leave behind hypotheses, experiments, findings, retros, and decision traces in a validated KB.

Feels like the next question after autoresearch is: not just “can the agent hill-climb?” but “can it accumulate taste + memory over weeks?”

https://t.co/TP6qkQPDwQ

[nick - @NeoHaskell ⏩ @NickSeagull]

On Thursday I'll be giving a talk
@AllDayDevOps
about

Bridging the AI to Ops Gap

I will cover some of the challenges that we've encountered on pushing AI to production @theagilemonkeys and how to mitigate them! #ai #LLMs #devops #coding

[Prof. Prokar Dasgupta OBE HonFREng @prokarurol]

Have been to many @ArcticMonkeys concerts but never heard of @theagilemonkeys until #ERUS24
Testing the @AppleVisionPro for computer vision and education in robotic surgery.
Kinder to the neck.
May enhance surgical performance.
Watch this space..

[Pablo Juárez DelDago @juarez_del_Dago]

Honored to be appointed as Chair of the new Technology Working Group of @ERUSrobotics. A sincere gratitude to the entire ERUS board, especially @AlbertoBreda1, for this incredible opportunity. Excited to collaborate with @pietrosdian, @BasileG_, and our talented team @theagilemonkeys. Together, we'll drive innovation and elevate our field! #ERUS24

[The Agile Monkeys @theagilemonkeys]

Spatial Computing. Time to dance.
 #AppleVisionPro

[The Agile Monkeys @theagilemonkeys]

It’s time to break free from Python in the AI world 👨‍💻⚔️🧌 Join the movement and contribute to building a new generation of production-ready tooling for AI in your programming language!
https://t.co/CLKChKlHC1

[nick - @NeoHaskell ⏩ @NickSeagull]

I've just created a non-technical guide for contributing to open-source projects on GitHub. Even correcting typos, changing colors, and fixing dead links counts as a contribution! https://t.co/YFKwk9xrVx

#GitHub #opensource

[The Agile Monkeys @theagilemonkeys]

Check out this guide to PrivateGPT and how to improve your company's knowledge management with AI while protecting security and privacy.
https://t.co/lZvnOdrE0Q

#PrivateGPT #AI #KnowledgeManagement #Security #Privacy

[Javi™ @javitm]

This is supposed to be a joke but… OMG, the amount of wasted time out there with the excuse of “doing agile” is unbelievable… agile was supposed to challenge the nonsense, but there’s so much nonsense built around it that the term no longer means anything…

[The Agile Monkeys @theagilemonkeys]

We’ve released a new issue of our newsletter. Join us and get inspired by one of our team members' learning journey with a side project, master the art of prompt engineering, acquire effective research paper reading skills, and more.
https://t.co/CTHqRvNsYA
#AI #innovation

[Javi™ @javitm]

At @theagilemonkeys, we've been discussing for a while the potential impact of Large Language Models (#llms) in the day-to-day work of #softwareengineers.

https://t.co/5VtFUybZTS

In this article, I share some opinions on how things could work out in the mid-term. A new wave of development tools based on LLMs will significantly increase our #productivity and acceptable #codequality.

Our role as engineers will fade away from code and require us to strengthen the other aspects of software engineering.

What do *you* think?

[The Agile Monkeys @theagilemonkeys]

An integral default approach for designing and implementing third-party integrations in event-sourced systems.
https://t.co/QnUzKH31nI
@boostthecloud #eventdriven #eventsourcing

[Javi™ @javitm]

If you’re open to use Serverless infrastructure (AWS or Azure), @boostthecloud provides nice abstractions to build event-driven workers on top of them.

We’re using a Booster app to handle a quite complicated consolidation process in a banking application among other things..

In the current version you have to implement jobs manually, using an entity to hold state and playing a bit with event handling to drive the workflow, but @JYCabello is planning to introduce a new Task abstraction that will make the process simpler 🙃

[Javi™ @javitm]

I just published an article exploring the close relationship we're finding between #eventsourcing and #AI and how relevant it might be for the future. Let me know what you think about it! https://t.co/I45ttusQA1