Inside Meltwater Engineering.

When our Elasticsearch cluster was recently overloaded, our API avoided more than 23,000 retries in 15 minutes, load that would otherwise have hit the struggling cluster. Here's how a simple token pool lets us retry when it helps and fail fast when it doesn't.

How we built small AI skills into Slack to measure contributor and reviewer quality, drive retros, and support feedback, while keeping judgment with the people doing the work.

Our team used AI-assisted development to rewrite two legacy applications: a decade-old CRUD app and an orphaned web service. Here is what worked, what didn't, and what we learned about the role of engineering judgment when AI is in the loop.

There's a particular kind of energy early in an engineering career. A quiet obsession. AI has brought that feeling back, not by replacing judgment, but by removing the friction that buried it.

How we overhauled Meltwater's speech-to-text pipeline with speech activity detection and dual-provider routing, cutting monthly model costs by 87% and making processing 1.5x faster.

We moved 1.23 TB of sentence embeddings out of our database cluster and into Amazon S3 Vectors, cutting the workload's run cost by roughly 90% and extending semantic search from 30 days to 24 months.

We rebuilt our journalist research pipeline from the ground up, splitting one generalist agent into focused specialist components. The result was 40% lower costs, around 10% higher prediction accuracy, and a system we can finally improve one piece at a time.

Keeping nearly 1,000,000 journalist profiles accurate used to be a fully manual job. We built an AI agent, Admiral, that cut research cost per profile by 85% and time per profile by 90%. Two lessons did most of the work: the value of a self-review step, and keeping the agent's toolset as...

The Meltwater Paris AI team attended the inaugural PyTorch Conference Europe at Station F. Here are our key takeaways on inference maturity, GPU kernel tooling, hardware co-evolution, and what it means for our AI platform.