Parseable

Let's talk rows and columns. Do you use or know the Parquet file format? In this -- Explain Parquet like I'm Five -- article, we dissect the file format, its benefits and uses, and why it's a great fit for logging systems like Parseable. dev.to/parseable/expl…

As volumes and use cases grow, data remains the moat in AI era.

Logs keep growing. Traditional observability stacks struggle. @parseablehq flips the model - building an observability data lake for agents, LLM workloads, and infra telemetry using an object-storage-first architecture on Tigris. tigrisdata.com/blog/case-stud…

It's been fun watching @parseablehq build an observability data lake on top of Tigris. They are basically enabling "AI-grade observability": model behavior, quality signals, token usage, and infra telemetry all in one place: tigrisdata.com/blog/case-stud…

We're looking for a growth and marketing expert to join our marketing team at Parseable, focused on developer and SRE audiences. You'll work on real world problems, shape how developers and SREs discover and adopt Parseable. What you'll do: - Content strategy and execution (blogs, social, community) - Demand generation and developer outreach - Positioning, messaging, and go-to-market experiments - Campaigns around launches, events, and partnerships Skills we're looking for: - Think and write independently - SEO and organic growth strategies - Community building and developer engagement - Email marketing and nurture campaigns - Tracking funnel metrics, attribution, and campaign performance This is a high-impact role at an early stage. You'll shape how the world discovers Parseable. 📍 In office (Whitefield, BLR) | ⌛ Full time Interested? email me at nitish [at] parseable [dot] com. #HiringAlert #marketing

Parseable is a telemetry datalake on object storage. Logs, metrics, traces, and AI agent telemetry go into one system, stored as open Parquet files on object store. You query with SQL or natural language. You keep data for 12 months instead of 30 days. You pay 80x lesser. Some context on why we built this. Observability costs grow linearly with data volume, but value doesn't. Most teams delete telemetry after a month because their platform makes retention irrational. AI workloads are making this worse — agents generate telemetry at extreme volume and cardinality, and that data is scattered across separate tools that don't correlate with each other. We spent the last year talking to engineering leaders about this. 75% cited cost as their primary observability concern. 84% of organizations plan to consolidate tools in 2026. The pattern was clear: teams are paying more each year, retaining less data, and getting slower queries. @parseablehq fixes the economics by fixing the architecture. Columnar Parquet format compresses telemetry up to 90%. Compute and storage scale independently. No pre-indexing, no write amplification, no JVM overhead. The entire engine is written in Rust and runs on @TigrisData , a globally distributed S3-compatible store with zero egress fees. Pro plan is $0.39/GB ingested. 12 months retention. All AI features included — natural language queries, forecasting, anomaly detection. Unlimited hosts, users, dashboards. 14-day free trial, no credit card. Enterprise gets bring-your-own-bucket for unlimited retention, data residency, BYOC or on-prem deployment, and Iceberg support. Self-hosted is still available. Single binary. Same engine. If you're running AI workloads alongside traditional infrastructure and tired of stitching together three tools to understand one incident, try out Parseable today at telemetry.new

Query performance: 𝗦𝗮𝗺𝗲 𝗹𝗲𝗮𝗴𝘂𝗲 𝗮𝘀 𝘁𝗵𝗲 𝗯𝗲𝘀𝘁 𝗢𝗟𝗔𝗣 𝗱𝗮𝘁𝗮𝗯𝗮𝘀𝗲𝘀. Parseable sits at ×2.97 relative query time, right alongside ClickHouse (×2.56) and DuckDB (×1.98). Being in the same league as analytical databases while also delivering the best compression and a full observability stack, that's best of both worlds. Full results: logg.ing/clickbench

We've been heads-down building Parseable core - the observability datalake for AI era. We recently benchmarked Parseable on ClickBench to understand the performance characteristics. Here are the results. Compression: #1 on ClickBench. 𝗣𝗮𝗿𝘀𝗲𝗮𝗯𝗹𝗲 𝗮𝗰𝗵𝗶𝗲𝘃𝗲𝘀 𝘁𝗵𝗲 𝗯𝗲𝘀𝘁 𝗰𝗼𝗺𝗽𝗿𝗲𝘀𝘀𝗶𝗼𝗻 𝗿𝗮𝘁𝗶𝗼 at 13.06 GiB, ahead of ClickHouse (13.73 GiB) and nearly 6× better than Elasticsearch (79.96 GiB). Compression is a design outcome. In observability, volumes drive the cost structurally. When you're ingesting petabytes of logs, metrics, and traces, the compression advantage compounds into massive savings. Best compression = most cost-effective solution. #Observability #LogManagement #ClickBench #DataEngineering #Parquet #CostOptimization #DevOps #SRE #Monitoring #Analytics

Finding the needle in a haystack of telemetry data? Filters are everything. Based on our users feedback, we just shipped new filters in Parseable to help you zero in faster. Check it out today at telemetry.new

New blog: Zero Code Observability for Kubernetes Services with OpenTelemetry parseable.com/blog/zero-code…

We ran @AnthropicAI's Opus 4.6 through 10 observability workflows using Parseable for log analysis, trace reconstruction, cascading failure RCA. It predicted OOM crashes, reconstructed 33-span traces, clustered 18 alerts into 3 root causes. parseable.com/blog/opus-4-6-…

We recently took some time off from building and shipping stuff to celebrate, recharge, and just enjoy each other's company outside of Slack threads and standups. Here's a little glimpse of Team Parseable out of office

What the heck is agent observability? Your AI agent can loop forever, burn through tokens, and confidently hallucinate, all while reporting "success." We broke down the 3 pillars you need to actually understand what's happening: 🔍 Evals 📊 LLM Monitoring 🛠️ Prompt Analysis parseable.com/blog/agent-obs…

It's now easier to see what's going on with LLM! The Broadcast feature from @openrouter sends traces straight to Parseable, giving you full visibility into token usage, costs, and latency across more than 200 models. Read more: parseable.com/blog/openroute…

Your coding agent just spent $2.40 and 3 minutes to add a print statement. The logs say "Task completed successfully." Sound familiar? We wrote a comprehensive guide on instrumenting 𝗱𝗶𝘀𝘁𝗿𝗶𝗯𝘂𝘁𝗲𝗱 𝘁𝗿𝗮𝗰𝗶𝗻𝗴 in coding agents with OpenTelemetry to capture every LLM call, tool execution, and token utilised. The guide includes a fully instrumented fork of SWE-agent (Princeton NLP's state-of-the-art coding agent) that sends traces to Parseable, where you can query everything with SQL. Check out the full guide: parseable.com/blog/tracing-c… #AIAgents #Observability #OpenTelemetry #DevTools #LLMOps

@LiteLLM Trace Analysis with Parseable We wrote a guide on sending LiteLLM traces to Parseable via OpenTelemetry. Whether you're tracking performance bottlenecks, analyzing token costs, or monitoring error rates across Claude, GPT-4, or Bedrock, this guide walks you through the entire pipeline. Perfect if you're building with LLMs and need to answer: Which model is slowest? Most expensive? Where are my bottlenecks? 15 minutes to set up. Covers latency analysis, cost tracking, error rates, and alerting. parseable.com/blog/litellm-t… #LLM #Observability #OpenTelemetry

"Is your observability bill skyrocketing? 📈 In our latest blog post, we cover how to route telemetry from applications to Parseable via @cribl_io stream to cut the noise and keep the signal. We dive into intelligent sampling, cost-effective storage, and anomaly detection using alerting. Read more: parseable.com/blog/cribl-par…

Introducing our new 𝗧𝗿𝗮𝗰𝗲 𝗩𝗶𝗲𝘄. We built this to give you complete visibility into the lifecycle of your requests, but it goes beyond just standard debugging. As we move toward 𝗔𝗴𝗲𝗻𝘁𝗶𝗰 𝗢𝗯𝘀𝗲𝗿𝘃𝗮𝗯𝗶𝗹𝗶𝘁𝘆, traces are becoming the critical "X-ray" for AI. They allow you to map the non-deterministic "thought process" of your agents, visualizing every decision node, tool use, and LLM interaction in a clear waterfall view.

𝗠𝗼𝗻𝗶𝘁𝗼𝗿 𝘆𝗼𝘂𝗿 𝗥𝗮𝘆 𝗰𝗹𝘂𝘀𝘁𝗲𝗿𝘀 𝘄𝗶𝘁𝗵 𝗣𝗮𝗿𝘀𝗲𝗮𝗯𝗹𝗲. Ray from @anyscalecompute powers some of the most complex AI workloads in production. But production clusters need historical context, centralized metrics, and the ability to correlate signals when things go wrong. We built a complete observability pipeline: Ray → Fluent Bit → Parseable ✅ Scrape Prometheus metrics from Ray ✅ Store them in OpenTelemetry format ✅ Query everything with SQL in Parseable 📖 Read the full guide: parseable.com/blog/monitorin…