One thing I built: defun https://github.com/killme2008/defun -- a macro for defining Clojure functions with pattern matching, Elixir-style. Still probably my favorite thing I've open sourced.

Comments URL: https://news.ycombinator.com/item?id=47581701

Points: 471

# Comments: 162

I built TMA1 because I wanted better understanding of what coding agents are actually doing — token/cost usage, tool calls, latency, failures, full session replays, etc. The most important part is that I don't want to sign up for or send data to any cloud service.

And of course it's opensourced:

https://github.com/tma1-ai/tma1

Comments URL: https://news.ycombinator.com/item?id=47474802

Points: 2

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47433475

Points: 12

# Comments: 0

Comments URL: https://news.ycombinator.com/item?id=47344855

Points: 3

# Comments: 0

Highlights:

Online region repartition (SPLIT/MERGE) to adjust partitions at runtime (distributed clusters; shared object storage + GC).

Faster Metric Engine PK filters via memcomparable byte comparisons (microbench: ~20–90× on dense codec).

PromQL planner prefers TSID (skips unnecessary label columns).

Would love feedback from folks running observability/time-series workloads, especially on upgrade paths + repartition behavior in real clusters. GitHub releases/changelog: https://github.com/GreptimeTeam/greptimedb/releases

In an agentic AI world this pressure gets worse: telemetry becomes more JSON-ish, more high-cardinality (tool names, model/version, prompt/template IDs, step graphs), and more bursty, so pre-modeling every metric up front breaks down faster.

The most surprising finding for me was the hash seed trick - using the same random seed across HashMaps in a two-phase aggregation gives you ~10% speedup on ClickBench. The bucket distribution from the first phase can be preserved during merge, eliminating rehashing overhead and making CPU cache happy.

We also discuss why Rust's prost library can be significantly slower than Go's protobuf implementation, and how fixing it improved our end-to-end throughput by 40%.

Happy to discuss Rust performance optimization or DataFusion internals.

Comments URL: https://news.ycombinator.com/item?id=46050757

Points: 2

# Comments: 0

We’ve been using Rust for about seven years now, and as long as you stay away from fancy unsafe tricks, you really can avoid most memory safety bugs.

PromQL, on the other hand, is purpose-built for observability — it’s optimized for time‑series data, streaming calculations, and real‑time aggregation. It’s definitely easier to learn and more straightforward when your goal is to reason about metrics and alerting.

SQL’s strengths are in relational joins, richer operator sets, and higher‑level abstraction, which make it more powerful for analytical use cases beyond monitoring. PromQL trades that flexibility for simplicity and immediacy — which is exactly what makes it great for monitoring.

Thank you for giving GreptimeDB a shout-out—it means a lot to us. We created GreptimeDB to simplify the observability data stack with an all-in-one database, and we’re glad to hear it’s been helpful.

OpenTelemetry-native is a requirement, not an option, for the new observability data stack. I believe otel-arrow (https://github.com/open-telemetry/otel-arrow) has strong future potential, and we are committed to supporting and improving it.

FYI: I think SQL is great for building everything—dashboards, alerting rules, and complex analytics—but PromQL still has unique value in the Prometheus ecosystem. To be transparent, GreptimeDB still has some performance issues with PromQL, which we’ll address before the 1.0 GA.

https://docs.greptime.com/reference/sql/functions/geo/#h3

These functions encode and decode latitude/longitude to H3 cells and provide utilities for querying cell properties, neighborhoods, distances, and relationships.

We agree that fine-grained access control is important. A read-only user role will be available in the next major release.

Comments URL: https://news.ycombinator.com/item?id=44176580

Points: 1

# Comments: 0