Achille

Gold dusted all we drank and ate

[Achille @__Achille__]

Are you in team light or dark mode?

Come by to meet @usefiretiger at @aiDotEngineer and claim your side!

[Achille @__Achille__]

Let’s get this started 🚀 @aiDotEngineer

[Achille @__Achille__]

@DanielLockyer @samlambert Some of us like to live dangerously

[Achille @__Achille__]

Friends don’t let friends write logs to their database

[Achille @__Achille__]

@davidlbowman Agreed

[Achille @__Achille__]

@samlambert 🙈

[Achille @__Achille__]

@rbranson Touché

[Achille @__Achille__]

All my recent tabs used to be Github, now it's filled of Google Cal and social media... But I'm still a software engineer, right? 😥

[Achille @__Achille__]

@perfloop https://t.co/FvkWo7JlIu

[Achille @__Achille__]

parquet-go merged its first PR from @perfloop 👏

An agent did the research, came up with an optimization opportunity, and opened a PR... no humans in the loop!

So much power is wasted to slow code, this is how we fix it.

[Achille @__Achille__]

When paired with the right context engineering, open-weight models have become good enough that they can match the quality of frontier models for certain tasks.

This is the story of how we migrated our core product from Claude to Deepseek ↓

[Rustam X. Lalkaka @lalkaka]

claude->deepseek: we cut token costs by 60%+ and maintained quality, all without a ton of eng effort.

a little story about our migration from claude to @deepseek_ai to power parts of @usefiretiger 's agent stack: https://t.co/fotB9xANhv

[Rustam X. Lalkaka @lalkaka]

driving continuous improvement loops in software systems is really hard, even if you have really smart agents (codex, claude, cursor), really smart humans, decent observability tools, etc.

today we're launching Service Monitors, loops in prod made easy. https://t.co/fV7KCtXpEp

[Jean-Denis Greze 💡 @jgreze]

Today, we’re launching @TownAI: the AI assistant that learns you.

We’re coming out of beta with a $55M Series A led by @ARampell at @a16z, with participation from @KirstenGreen at @forerunnervc and continued support from @firstround, @altcap, and @conviction.

Right now, getting real value from AI means prompting, configuring, building workflows, managing agents.

We think that’s backwards.

The future of AI is a companion that already knows you and how you work. Town connects across your inbox, calendar, Slack, docs, messages, and workflows to understand what you need, then starts doing the work with you.

Drafting. Scheduling. Project tracking. Follow-ups. Context gathering. Multi-step tasks. And it only acts when you say so.

All adapting to your voice, priorities, routines, and relationships over time.

Your Townie is the AI assistant you actually need.

[Achille @__Achille__]

@lalkaka took some time to chat with them to learn more about it, you can read the full story here https://t.co/0bx3weAHJQ

[Achille @__Achille__]

@usekernel is building the browser infrastructure for agents, and this comes with a whole new set of reliability challenges.

[Achille @__Achille__]

To respond to those reliability challenges, they've adopted @usefiretiger very early on, and were amongst the first companies to start using our change monitor agents in their development and release workflow.

[Achille @__Achille__]

@LokeshJambu @boristane @o11ycon Hopefully cost(maintenance) != 1/cost(development) 😅

[Achille @__Achille__]

@boristane's closing question at @o11ycon

"If the cost of code is zero, what is the value?"

I'm glad it's a long weekend because I have a lot to reflect on...

[Achille @__Achille__]

"With AI adoption, Build got ahead of Operate, we're here to fix that"

[Achille @__Achille__]

Really nice looking stage today @o11ycon

[Achille @__Achille__]

The DORA manifesto
https://t.co/QaB4LemsbE