Sillage
B2B platform · AI agentsFrom February to May 2026, I joined Sillage (a B2B sales-signals platform powered by AI agents) for a 3-month mission as a software engineer, in a 3-person engineering team. I architected a multi-provider deep-research pipeline with Claude as the fit-scoring layer (a full signals schema plus a Strapi data model), and designed an automated bug-fixing pipeline connecting Claude Code agents to Notion, GitHub, Linear and Slack to generate AI-reviewed pull requests.
Detecting the right buying signals and prioritizing the accounts to target, by hand, didn't scale.
A multi-provider deep-research pipeline with Claude for fit scoring, and an automated bug-fixing pipeline (Claude Code agents × Notion, GitHub, Linear, Slack).
Continuously qualified accounts and AI-reviewed pull requests, shipped faster.
In pictures
Software engineeringAI agentsA similar case at your company?
Let's book 30 minutes to talk it through. Concrete, no commitment.
More case studies
About 3h per client per week of manual compilation gone: the team steers prices in real time instead of re-keying numbers.
View case →
Hundreds of applications handled frictionlessly, a faster selection.
View case →
An experience as simple as a regular app, without giving up any blockchain security.
View case →
The first exhibition sold as NFTs in Upper Savoy, widely picked up by the regional press.
View case →
Scalable processes, ready to take on the volume.
View case →