
Building @fond_kitchen, a structured, multilingual ingredient brain for home cooks. Also building trading stuff at @Lynx_Protocol
Remote
Joined July 2021
- Tweets8.7K
- Following723
- Followers2.5K
- Likes11.8K
Did you see that Paper shipped support for Cloud Agents this week?
Check out this tutorial from @harcourttt on how to get the most from Paper's remote MCP
Design from any agent, any place, any time
Bastien retweeted
Quad9's forced withdrawal from France highlights the dangers of mandating DNS-level blocking.
Read more: cfl.re/4f0cowv
For years I had UI ideas I couldn’t ship.
Not because I couldn’t code them. Because iterating on them was too slow.
I always knew how an interface should feel. But between sketching, mocking, coding and tweaking, most ideas died at version two. I never learned Figma properly, so the gap between what I imagined and what I built stayed wide.
That gap is mostly gone now. My setup:
→ @paper for the canvas. My coding agent can read and write to it, and since it’s built on web standards, what I design is close to what ships.
→ @impeccable_ai for the critique. It catches what I’d miss: spacing, hierarchy, contrast, inconsistent components.
What changed is the cost of a bad idea. I can try five directions in the time one used to take, throw away four, and turn the survivor into a design system: tokens, components, rules the agent follows next time.
On @fond_kitchen - The Cooking App, it took me a lot of iterations to figure out what I actually had in mind. Before, I’d have stopped at version two and shipped something close enough. This time I kept going until it felt right.
Same with animated SVGs. They were always the part I cut for time. Since Claude Opus 5.5, they’re cheap enough to try, so they finally make it in.
The honest limit: AI makes generic UI very cheap. If you don’t know what good looks like, you’ll ship the same clean, forgettable interface as everyone else.
Taste is the bottleneck now, not tooling.
What’s the one step in your UI process you still can’t speed up?
Bastien retweeted
Our project manager is a markdown file in our repo.
Our coding agents create and update our @linear issues through MCP, following a few rules we wrote down once. Nobody moves a status by hand anymore.
The whole setup for two co-founders:
Cycles: off.
We're pre-launch. Sprints would just be a deadline we set to disappoint ourselves. Work lives in projects, one per workstream, each with a one-sentence summary.
Triage: we're our own first users.
The app has a feedback button. Each report becomes a Linear issue, then an agent picks it up and opens a PR with the fix. When real users arrive, the loop will already work.
Code review: also in Linear.
The diff sits next to the issue. We barely open GitHub anymore.
Labels: one of each per issue, plus a tag.
→ Type: bug, feature, improvement, chore
→ Area: API, web, mobile, growth, operations
→ Product area: recipes, grocery lists, meal planning…
→ Tag: User feedback
Weekly recap: Linear summarizes what shipped. We start our catch-up from there.
The rules in AGENTS.md:
→ Every PR has an issue
→ Search before creating a new one
→ Titles are plain sentences about the outcome
→ PRs end with "Fixes FON-123", so Linear moves the issue itself: In Progress on open, Done on merge
No status updates, no dragging cards.
Small teams: what did you remove from your setup that you never missed?
Bastien retweeted
Your first agent doesn't need to be complex or scary. Skip the chatbot and build one that tells you every morning whether your site is growing.
Ours is called Boromir. It runs daily on a @Cloudflare Worker and posts a short report on fond.kitchen's growth in Slack.
Here's how it works, in case you want to copy it:
1. Put the data behind MCP.
We have a small internal dashboard, Sauron (yes, there's a theme), that exposes read-only tools over MCP: one for @posthog (visitors and waitlist signups) and one for Google Search Console and Bing Webmaster (clicks, impressions, queries).
Read-only matters. The agent can look at everything and break nothing.
2. Let code do the maths.
The code pulls yesterday's numbers and computes every comparison. The model only gets the finished numbers and writes the interpretation. It never queries data or calculates anything itself, so it can't invent a figure.
3. Ask for changes, not dashboards.
Every morning it reports visits, waitlist signups, top pages and traffic sources, compared with the same weekday over the previous four weeks. Weekly and monthly reports show the trends.
4. Let it say "nothing to do".
Each report can end with up to three things worth checking. Most days the right answer is none, and the agent is told an empty list is fine.
5. Read it where you already are.
The daily report lands in Slack. And when I want search data, I mention Boromir in the thread and it pulls Search Console and Bing on the spot.
It costs almost nothing to run. Cloudflare hosts it, and all you need is an @OpenRouterAI key for the model. The agent itself is one prompt and a couple of tools (@flueai).
The honest limit: false positives. Early on, Boromir kept suggesting the same "fix" every day even when it didn't apply. The agent only knows what you tell it. Once we gave it more context about our goals and what we'd already ruled out, the repeated suggestions stopped.
What would you want your first agent to report on every morning?
Bastien retweeted
Hey X 👋
My current role is coming to an end, so I'm looking for my next one: Senior Product Engineer, full remote from France
13 years building products, 10 of them remote, and 8 of those spent living in Asia, so async and time zones don't scare me. Mostly React, TypeScript and Node, plus Android/Kotlin/KMM.
Some recent stuff:
At Lynx Finance I rebuilt the perps trading app (over $5B in volume), then built a multiplayer arcade game where people play for real money. The fun part was making sure the money never breaks.
Before that I ran tech and product at Owned, a Counter-Strike tournament platform, including fully automated on-chain prize distribution: more than $200k paid out to players.
And on the side, I'm building my dream cooking app, Fond @fond_kitchen , with my co-founder, because when I'm not building, I'm cooking. At its core is an AI ingredients engine we built together: you give it any recipe, from a website, a video or a messy blog post, and it turns it into clean structured data, with every ingredient, quantity and unit understood. Harder than it sounds. I lead product, growth and the apps. Not launched yet, but already 900 people on the waitlist and around 800 visitors a day from Google.
These days I think a lot about how teams should work in the AI era: tools like Linear and PostHog, good internal tools, and nice UI / MCP everywhere. I like small teams where engineers own the product, not just the tickets. Food, consumer apps, fintech and trading are my favourite playgrounds. But really just looking for a neat product and team.
If you're hiring or know someone who is, my DMs are open. A share helps a lot too!
Hey X 👋
My current role is coming to an end, so I'm looking for my next one: Senior Product Engineer, full remote from France
13 years building products, 10 of them remote, and 8 of those spent living in Asia, so async and time zones don't scare me. Mostly React, TypeScript and Node, plus Android/Kotlin/KMM.
Some recent stuff:
At Lynx Finance I rebuilt the perps trading app (over $5B in volume), then built a multiplayer arcade game where people play for real money. The fun part was making sure the money never breaks.
Before that I ran tech and product at Owned, a Counter-Strike tournament platform, including fully automated on-chain prize distribution: more than $200k paid out to players.
And on the side, I'm building my dream cooking app, Fond @fond_kitchen , with my co-founder, because when I'm not building, I'm cooking. At its core is an AI ingredients engine we built together: you give it any recipe, from a website, a video or a messy blog post, and it turns it into clean structured data, with every ingredient, quantity and unit understood. Harder than it sounds. I lead product, growth and the apps. Not launched yet, but already 900 people on the waitlist and around 800 visitors a day from Google.
These days I think a lot about how teams should work in the AI era: tools like Linear and PostHog, good internal tools, and nice UI / MCP everywhere. I like small teams where engineers own the product, not just the tickets. Food, consumer apps, fintech and trading are my favourite playgrounds. But really just looking for a neat product and team.
If you're hiring or know someone who is, my DMs are open. A share helps a lot too!
Your first agent doesn't need to be complex or scary. Skip the chatbot and build one that tells you every morning whether your site is growing.
Ours is called Boromir. It runs daily on a @Cloudflare Worker and posts a short report on fond.kitchen's growth in Slack.
Here's how it works, in case you want to copy it:
1. Put the data behind MCP.
We have a small internal dashboard, Sauron (yes, there's a theme), that exposes read-only tools over MCP: one for @posthog (visitors and waitlist signups) and one for Google Search Console and Bing Webmaster (clicks, impressions, queries).
Read-only matters. The agent can look at everything and break nothing.
2. Let code do the maths.
The code pulls yesterday's numbers and computes every comparison. The model only gets the finished numbers and writes the interpretation. It never queries data or calculates anything itself, so it can't invent a figure.
3. Ask for changes, not dashboards.
Every morning it reports visits, waitlist signups, top pages and traffic sources, compared with the same weekday over the previous four weeks. Weekly and monthly reports show the trends.
4. Let it say "nothing to do".
Each report can end with up to three things worth checking. Most days the right answer is none, and the agent is told an empty list is fine.
5. Read it where you already are.
The daily report lands in Slack. And when I want search data, I mention Boromir in the thread and it pulls Search Console and Bing on the spot.
It costs almost nothing to run. Cloudflare hosts it, and all you need is an @OpenRouterAI key for the model. The agent itself is one prompt and a couple of tools (@flueai).
The honest limit: false positives. Early on, Boromir kept suggesting the same "fix" every day even when it didn't apply. The agent only knows what you tell it. Once we gave it more context about our goals and what we'd already ruled out, the repeated suggestions stopped.
What would you want your first agent to report on every morning?
Our project manager is a markdown file in our repo.
Our coding agents create and update our @linear issues through MCP, following a few rules we wrote down once. Nobody moves a status by hand anymore.
The whole setup for two co-founders:
Cycles: off.
We're pre-launch. Sprints would just be a deadline we set to disappoint ourselves. Work lives in projects, one per workstream, each with a one-sentence summary.
Triage: we're our own first users.
The app has a feedback button. Each report becomes a Linear issue, then an agent picks it up and opens a PR with the fix. When real users arrive, the loop will already work.
Code review: also in Linear.
The diff sits next to the issue. We barely open GitHub anymore.
Labels: one of each per issue, plus a tag.
→ Type: bug, feature, improvement, chore
→ Area: API, web, mobile, growth, operations
→ Product area: recipes, grocery lists, meal planning…
→ Tag: User feedback
Weekly recap: Linear summarizes what shipped. We start our catch-up from there.
The rules in AGENTS.md:
→ Every PR has an issue
→ Search before creating a new one
→ Titles are plain sentences about the outcome
→ PRs end with "Fixes FON-123", so Linear moves the issue itself: In Progress on open, Done on merge
No status updates, no dragging cards.
Small teams: what did you remove from your setup that you never missed?
Bastien retweeted
How Gridy works: pick a square ahead of the ETH price. If the price touches it while its column is live, you're paid that square's multiplier.
Squares start at 0.1 USDC.
app.lynx.finance/gridy?utm_s…
Playing Gridy on @Lynx_Protocol
Trading made fun, available on @BNBCHAIN
Bastien retweeted
first piece of our @linear company documents project just landed in private beta...
(your documents have never been this fast, or looked this good by default)
Looking for a fun trading game?
We released Gridy at @Lynx_Protocol, along with other games.
The new weekly leaderboard is up.
Available on BSC, Sonic & Flare
app.lynx.finance?referrer=0x…
Bastien retweeted
People keep asking how companies like Ramp, Linear, and Anthropic ship so much software with agents.
There is no new SDLC or secret set of tools. The tools are increasingly available to everyone. You can build your own, use Linear, Claude, or whatever works.
The bigger difference is ambition. These companies encourage their teams to build more, experiment more, and let agents do more of the work. That's the main different between "AI native" or not. You try to use the technology to the max.
But don't get there by adopting a tool. You get there by changing what you believe your teams are capable of.
So in the very simplest way:
- let agents build
- make it easier and faster to contribute
- increase the ambition of what your teams attempt.
In practice, that means things like:
- Standardize around cloud agents. Most of our code (5M lines last quater) is now written in Linear with our Linear cloud coding agent. Cloud sandboxes make standardizing the system and environment easier for everyone and enables faster contributions
- Linear agent (with code access and cloud agent) operates in Slack too. If there is a discussion or someone notices a problem just say "@linear fix this" (decrease the time from something noticed to fixed)
- As bugs, errors etc arrive in Linear, send them to Triage, and have Loop to auto-investigate (connect to tools like Sentry, Datadog), and write a fix
- Let designers, PMs, non-technical people contribute code where useful. If they notice a problem, tell them they can submit a fix. If they want to try something, ask them to build a prototype.
- Set up code review agent and system to make sure PRs are safe to merge.
- Run security as an automation, have agents constantly try to find security issues in your codebase, as they find, send it to a coding agent
- Instead of working on issue level, increase ambition, make a whole project, ask the agent to ask questions, investigate code base, create issues and have the agent work on the whole project at once. Set up goals like "give me proof that what you build works". All our PRs now come with screenshots and videos and preview build you can
- Bring customer feedback to the system like Linear, and make it available for your team. At Linear every project now starts with 10-500 customer notes that influence the direction of the work, and which customers we should ask to test the solution first.
- Set up Loops/automations to constantly update reference documents for agents, like product area/code owners so work can be routed to right humans
Then what is left is how do you set direction for your teams (initiatives, projects they should work on) and how do you review results (better code review, Linear does it with Guided Reviews), how you maintain quality/cohesion which is not a new problem either.
Basically most of the problems you always had remain, but the execution bit is now more and more solved.
At @linear we can help with automation, cloud agents, and code review, customer feedback and planning bits.
But we can't change your mindset or how your company operates.
If your company is still spending energy debating jira workflow steps, you're probably solving the wrong problems.
The US is about to join an EU innovation program.
BREAKING: 🇺🇸 Democrats face a push for a “day 1” ban on AI superintelligence, with up to 20 YEARS in prison for violators.
Arizona Rep. Yassamin Ansari says a pause on advanced AI and a ban on artificial superintelligence must be “top of the Democratic agenda on day 1.”
A bill that would do exactly that was introduced by Bernie Sanders and Greg Casar in September.
Under the bill:
- Developing or deploying superintelligent AI would be banned
- Advanced AI development would pause until a new federal regulator sets rules and a model review process
- A new cabinet-level AI department would monitor frontier models and oversee removing dangerous capabilities, like ignoring shutdown commands
- Companies that try to get around it would face the “corporate death penalty”
- Individuals would face up to 20 years in prison
AI and data centers were a “crucial or major factor” for 56% of voters in the midterms, per an NBC News exit poll.
54
Announcing a big fundamental ship in Paper today
Show us what you design!
Re-made my local wine shop website to help them a bit.
It was fun. Fully made using @EmDashCMS so they can magically edit content with Claude MCP.
Shop with @vendure_io
REALLY fun stack to use.
cavetourneegenerale.com/















