---
title: "Latest reads"
description: "The engineering internet, summarised so you can actually read it."
---

# Latest reads
> The engineering internet, summarised so you can actually read it.

## Articles

### [How we built OCA, our AI on-call assistant](https://yomu.fyi/post/how-we-built-oca-our-ai-on-call-assistant.md)
- Company: [Ramp](https://yomu.fyi/company/ramp.md)
- Author: Richard Marmorstein, Austin Ray
- Published: Sep 9, 2026
- Type: Explainer

Ramp built OCA, an AI on-call assistant that investigates incidents, shares findings in Slack, answers follow-up questions, and can ask Inspect to prepare code-fix pull requests for human review. In the past four weeks, it assisted with 575 of 1,220 incident-related merged PRs; among incidents with merged fixes, OCA-assisted fixes averaged 37% fewer engineers and about 50% less estimated engineer time, based on Slack activity. Investigations run in a monorepo through Claude Code or Codex, with Temporal managing workflow, while prompts, custom tools, skills, and six one-shot hooks provide successive evidence and reasoning checks. Session-driven testing exposed errors such as misleading one-hour log windows, leading to just-in-time instructions and interim updates that balance accuracy with response speed. OCA remains human-in-the-loop: it has broad read-only access, narrowly scoped writes, and cannot merge changes, although the team is building toward end-to-end routine incident handling.


### [The Evolution of WSL Support in JetBrains IDEs](https://yomu.fyi/post/the-evolution-of-wsl-support-in-jetbrains-ides.md)
- Company: [Jetbrains](https://yomu.fyi/company/jetbrains.md)
- Author: Kristina Pchelintseva
- Published: Sep 9, 2026
- Type: Explainer

JetBrains IDEs have supported WSL through several architectures, but the 2026.2 release makes Native mode the recommended entry point for projects stored there. In Native mode, the Windows IDE delegates filesystem and process operations to the Rust-based IJent agent inside WSL, with EelApi abstracting local and remote environments for IDE and plugin code. Earlier approaches included 9P filesystem access, WSLg, and Remote Development, with problems involving symlink handling, latency, GUI limitations, installation overhead, and client–backend communication costs. Performance tests on a cold first open of spring-framework measured Native mode at 11.5 seconds to become ready versus 18.5 seconds for 9P, while project-tree scanning took 3.7 versus 8.1 seconds. Remote Development remains available, but direct project opening is now preferred in IntelliJ IDEA, WebStorm, and PhpStorm.


### [Supabase is now available in Gemini Enterprise](https://yomu.fyi/post/supabase-is-now-available-in-gemini-enterprise.md)
- Company: [Supabase](https://yomu.fyi/company/supabase.md)
- Author: shaziya\_bandukia, elsa\_heffernan
- Published: Sep 9, 2026
- Type: Announcement

Supabase is now available as a prebuilt connector in Google Cloud Gemini Enterprise, allowing teams to query and take action on Supabase projects through natural-language prompts. After an organization-level connection, Gemini Enterprise calls the connector for live answers, does not store or index Supabase data, and can combine results with connected tools such as GitHub, Linear, Notion, Slack, and Salesforce. The integration supports questions about tables, database metrics, and table structure, while exposed tools are marked read-only or destructive. Business users connect through the connector menu, whereas Standard, Plus, and Frontline administrators use a client ID and secret in the admin console; the connector is available across all editions.


### [Personalize your product's text-to-speech voice for any language: Fine-tuning with Kubeflow Trainer on Red Hat OpenShift AI](https://yomu.fyi/post/personalize-your-product-s-text-to-speech-voice-for-any-language-fine.md)
- Company: [Red Hat](https://yomu.fyi/company/red-hat.md)
- Author: Dmytro Hryshchenko, Abhijeet Dhumal
- Published: Sep 9, 2026
- Type: Tutorial

Training an English-centered text-to-speech model for Turkish can produce fluent but hard-to-understand pronunciation for native speakers. This tutorial fine-tunes unsloth/orpheus-3b-0.1-pretrained with about 81,000 Turkish text-audio pairs on Red Hat OpenShift AI using a Kubeflow Trainer TrainJob distributed across two nodes with two A100 GPUs each. LoRA freezes the 3-billion-parameter base and trains 97.3M adapter parameters, while preprocessing converts text to Llama-3 tokenizer tokens and audio to three-layer SNAC codes, masking text labels during scoring. The merged model is evaluated against the base on ten Turkish sentences using Whisper transcription and WER/CER metrics. Mean WER falls from 2.44 to 0.35 and CER from 1.59 to 0.09, though the authors describe the small test as directional.


### [Evaluation-First AI Agents: How Zepto Scales Customer Support on Databricks and MLflow](https://yomu.fyi/post/evaluation-first-ai-agents-how-zepto-scales-customer-support-on-databr.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Gireesh Sreedhar KP, Deepak Dhankani, Eash Sharma
- Published: Sep 9, 2026
- Type: Explainer

Zepto describes its customer-support system, processing more than 100,000 tickets daily, and the assurance problems that emerged as volume, categories, and customer behavior changed. Its response is an evaluation-first architecture on Databricks and MLflow, using development and production loops connected by a quality gate. Agent traces capture prompts, completions, retrieved documents, tool calls, latencies, and decision paths through MLflow, OpenTelemetry spans, and Unity Catalog-backed Delta tables; stakeholders define evaluation pillars and gates around a living golden dataset. The dataset grew from 500 examples with an eight-point development–production accuracy gap to 5,247 examples with a 0.4-point gap over six months, while prompt optimization and multiple scorers support iteration. Reported outcomes include 80% plus of tickets fully managed with human oversight, a 65% reduction in support cost or tickets, 20% higher CSAT, 8% better accuracy, and faster development and resolution.


### [Benchmark Shows MongoDB Atlas Manages Thousands of Payments Per Second with Zero Data Loss](https://yomu.fyi/post/benchmark-shows-mongodb-atlas-manages-thousands-of-payments-per-second.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Published: Sep 9, 2026
- Type: Benchmark

A benchmark of the Icon Payments Framework (IPF), trusted by Citi, NatWest, and BNP Paribas, on MongoDB Atlas demonstrated extreme performance and resilience for instant payment processing. The system sustained 6,000 end-to-end payments per second, requiring approximately 430,000 simultaneous database operations, with a mean end-to-end latency of 0.51 seconds and persistence latency below 30 milliseconds. Upgrading from MongoDB Atlas 8.0 to 8.3 reduced persistence latency by roughly 50%, from 40 to 20 milliseconds at 4,000 payments per second, and throughput scaled near-linearly across the 500 to 6,000 range while latency remained flat until peak and hot-shard CPU utilization sat at about half. During a primary failover, a new node was elected in five seconds with zero data loss and no application restart. An application node failure recovered to 3,500 transactions per second in 90 seconds with zero data loss.


### [Build a Fraud Review Agent with MongoDB Atlas and Claude Managed Agents](https://yomu.fyi/post/build-a-fraud-review-agent-with-mongodb-atlas-and-claude-managed-agent.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Published: Sep 9, 2026
- Type: Tutorial

Claude Managed Agents runs Anthropic’s Claude as a hosted agent without managing inference infrastructure. Pairing it with MongoDB Atlas provides fast access to operational and vector data through a single pymongo connection, eliminating the need for separate vector, full-text, and graph stores. This tutorial builds a human-in-the-loop fraud review agent that retrieves flagged transactions, runs four retrieval patterns (vector, full-text, hybrid, and graph) against one collection, and escalates risky cases to a human reviewer using Managed Agents’ native pause. The credential-safe integration uses host-side custom tools so the MongoDB connection string never enters the agent sandbox, keeping secrets on your side. The agent verifies a signed authorization mandate, uses $graphLookup to detect mule chains and $rankFusion for hybrid search, writes immutable verdicts and audit events back to the same cluster, and triggers durable human pauses on structuring deposits just under $5,000, wires over $50,000, and ring signals from graph traversal.


### [The Open Source AI Stack](https://yomu.fyi/post/the-open-source-ai-stack.md)
- Company: [Together AI](https://yomu.fyi/company/together-ai.md)
- Published: Sep 9, 2026
- Type: Explainer

The post presents an open model AI stack for developers moving from closed-source models, focusing on ownership, control, and economics. It organizes the stack into independent layers: model, inference, gateways and routers, harness, and tools such as Skills and MCP. Models generate responses from prompts; large models generally handle ambiguity, multi-step reasoning, and long context better, while small models can be faster and cheaper for tightly specified work. The separation lets developers choose providers and models independently and experiment with newly released models without rebuilding their workflow, while the post uses Kimi K3 and GLM 5.3 Flash to illustrate differences in active parameters, capability, speed, and cost.


### [10+ best practices to improve your email deliverability](https://yomu.fyi/post/10-best-practices-to-improve-your-email-deliverability.md)
- Company: [Twilio](https://yomu.fyi/company/twilio.md)
- Author: Jesse Sumrak
- Published: Sep 9, 2026
- Type: Tutorial

Email deliverability determines whether messages reach inboxes rather than merely being accepted by recipient mail servers, and the guide distinguishes delivery rate from inbox placement. It recommends authenticating domains with SPF and DKIM, using appropriate IP allocation and warm-up, applying double opt-in, writing genuine subject lines, offering preference controls, cleaning lists, avoiding purchased addresses and spam traps, sending valuable content, and monitoring metrics. The stated benchmark for good deliverability is at least 95%, while the excerpt cites Twilio SendGrid's 99% API rate and warns that changing providers does not repair domain or IP reputation. Case studies report Shopify at 99.5% overall delivered and 91.3% inbox placement, and Klaviyo at 99.94% average delivery, with the latter also reporting 1.9-second median delivery speed.


### [From AI Agent to Human Agent: ElevenLabs Call Handoffs with Twilio](https://yomu.fyi/post/from-ai-agent-to-human-agent-elevenlabs-call-handoffs-with-twilio.md)
- Company: [Twilio](https://yomu.fyi/company/twilio.md)
- Author: Ricardo Bangueses
- Published: Sep 9, 2026
- Type: Tutorial

This reference integration connects an ElevenLabs voice agent to Twilio Programmable Voice, enabling escalation from automated support to a human when a conversation becomes complex, sensitive, or unresolved. Twilio retains ownership of the original inbound call while ElevenLabs conducts the real-time conversation; an escalate\_to\_human webhook tells Twilio to update that parent Call SID with new TwiML. In the Flex example, Enqueue sends the caller into a TaskRouter workflow, creating a voice task with intent and summary context, while the Call SID and metadata such as direction and phone numbers travel into ElevenLabs as dynamic variables for the handoff. The blueprint describes Studio-owned routing, direct TaskRouter or Flex routing, and custom TwiML destinations; Conversation Memory and Conversation Intelligence are optional additions, and the implementation is explicitly a proof of concept rather than a production drop-in.


### [The Anatomy of Harness Engineering: How to Evaluate, Iterate, and Guard AI Coding Agents](https://yomu.fyi/post/the-anatomy-of-harness-engineering-how-to-evaluate-iterate-and-guard-a.md)
- Company: [Google](https://yomu.fyi/company/google.md)
- Author: Taylor Mullen, Christian Gunderman
- Published: Sep 9, 2026
- Type: Explainer

The post argues that end-to-end benchmarks such as Terminal-Bench and DeepSWE show overall agent performance but rarely explain why behavior changes. It presents behavioral evaluations as fast, deterministic, unit-style checks of observable intermediate actions, such as asking clarifying questions, running validators, or invoking web search instead of guessing. The proposed workflow starts with a recent failure, uses strict milestone assertions for simple tasks and flexible outcome-based or LLM-as-a-judge checks for complex tasks, then runs batch evaluations to track aggregate pass rates despite model nondeterminism. These evaluations are positioned for the second phase of development, after an agent can dogfood its codebase, and as complementary to macro benchmarks: behavioral tests enable prompt, tool-schema, and model iteration while guarding against regressions.


### [Announcing ADK for Kotlin 1.0: Building Production-Ready AI Agents in Kotlin, Android, and Beyond](https://yomu.fyi/post/announcing-adk-for-kotlin-1-0-building-production-ready-ai-agents-in-k.md)
- Company: [Google](https://yomu.fyi/company/google.md)
- Author: Guillaume Laforge
- Published: Sep 9, 2026
- Type: Announcement

ADK for Kotlin 1.0 is announced as a general availability release for building AI agents with Kotlin across Android, JVM server applications, and Java-integrated systems. The release targets a production-ready toolkit while preserving an idiomatic, lightweight, composable approach, and it reaches feature parity with ADK 1.0 Core, including hierarchical agents, context compaction, human-in-the-loop confirmation, resumable sessions, annotated tools, and Java interoperability. A Kotlin Multiplatform core stays agnostic to model backends, session providers, and memory systems, while Android extensions support LiteRT-LM, ML Kit (beta), Firebase AI Logic, Room, AppSearch, and file artifacts. The database incident example combines compile-time KSP-generated function schemas, suspend-capable @Tool functions, and dynamically loaded SKILL.md playbooks with progressive disclosure. The release provides 1.0.0 dependencies, documentation, a repository, and sample applications.


### [How Metadata and Runtime Telemetry Reveal Which Enterprise System Problems to Fix First](https://yomu.fyi/post/how-metadata-and-runtime-telemetry-reveal-which-enterprise-system-prob.md)
- Company: [Salesforce](https://yomu.fyi/company/salesforce.md)
- Author: Scott Nyberg
- Published: Sep 8, 2026
- Type: Explainer

Salesforce Health Insights addresses enterprise org health monitoring by combining configuration metadata with runtime telemetry to identify and prioritize remediation. The team evolved a static report into an in-app experience with durable finding dispositions, approximately 400+ signals across domains, and standards informed by Well-Architected principles and implementation anti-patterns. A shared harmonization layer reconciles fragmented datasets, connector patterns link runtime activity to metadata components, and the system preserves data origin, freshness, and access requirements while keeping deterministic findings distinct from externally supplied context. Prioritization combines findings with request volume, application and database CPU, call behavior, and peak-business-hour traffic to rank next best actions by criticality and effort. A headless JSON schema supports interfaces including an MCP connection, while the deterministic core is intended to produce dependable recommendations and extend access beyond the primary interface.


### [Ceilometer: Uber’s Adaptive Benchmarking Framework](https://yomu.fyi/post/ceilometer-uber-s-adaptive-benchmarking-framework.md)
- Company: [Uber](https://yomu.fyi/company/uber.md)
- Published: Sep 8, 2026
- Type: Explainer

Ceilometer is Uber’s adaptive benchmarking framework for qualifying new server types, cloud SKUs, kernel upgrades, configuration changes, and machine-learning frameworks before production rollout. It replaces manual, fragmented testing with standardized configurations, automated cluster bring-up and service deployment, portable containerized harnesses, and reporting that includes server specifications, reference SLAs, normalization, resource utilization, cost efficiency, and reliability signals. Its distributed architecture runs synthetic and production-like workloads across dedicated machines, processes and enriches raw results, stores them in blob storage and a centralized data warehouse, and exposes them through an analytics service. The framework covers CPU and memory benchmarks, tools such as SpecCPU2017, SPECjbb2015, NetPerf, and FIO, plus stateful, stateless, batch, and ML workloads; Uber reports uses including regression detection, database validation, and canary testing of external upgrades.


### [SQL Data Types: Reference and Best Practices](https://yomu.fyi/post/sql-data-types-reference-and-best-practices.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Databricks Staff
- Published: Sep 8, 2026
- Type: Explainer

SQL data types define what values columns accept, how those values are stored, and how databases process them, making type selection central to schema design. The reference covers numeric, character, date/time, and specialized types, noting that MySQL, PostgreSQL, SQL Server, and Oracle differ in naming, precision, limits, and storage requirements. It recommends choosing the smallest type that safely accommodates expected values, checking range, precision, NULL requirements, string length, and time-zone needs before deployment. Examples contrast CHAR with VARCHAR, DECIMAL with FLOAT, DATE with TIMESTAMP, and integer sizes such as TINYINT, INT, and BIGINT; indexes and distributed engines also benefit from right-sized data. Sample data, execution-plan analysis, storage measurements, schema audits, and documentation help validate choices, identify waste or overflow risks, and preserve reasoning behind edge cases.


### [Build durable agents with Temporal and Lakebase](https://yomu.fyi/post/build-durable-agents-with-temporal-and-lakebase.md)
- Company: [Databricks](https://yomu.fyi/company/databricks.md)
- Author: Sam Ingbar
- Published: Sep 8, 2026
- Type: Explainer

Long-running personal-loan underwriting agents must preserve completed work, resume after worker failures, tolerate repeated tool calls, wait for reviewers, expose operational state, and retain an audit trail. The reference implementation combines Temporal durable Workflows and Activities with Lakebase Postgres, using Temporal Event History for replay and Lakebase schemas for run status, evidence, review state, messages, and metrics. Deterministic identifiers, constraints, guarded updates, and Postgres upserts make repeated Activity attempts target the same logical records, while Signals carry human decisions and long waits do not occupy workers. A Unity Catalog synced table supplies governed underwriting policy, and optional Change Data Feed publishes operational changes to Delta history tables. The implementation includes a React/FastAPI interface, mocked deterministic providers, OAuth credential refresh, and 21 passing tests, but does not validate production lending, compliance, security, scale, or an observed end-to-end feed.


### [Introducing chdb Postgres extension: High-performance imports from cloud storage](https://yomu.fyi/post/introducing-chdb-postgres-extension-high-performance-imports-from-clou.md)
- Company: [Clickhouse](https://yomu.fyi/company/clickhouse.md)
- Published: Sep 8, 2026
- Type: Announcement

The new chdb Postgres extension uses the chDB library, an in-process ClickHouse engine, to import and export data between Postgres and files on cloud storage. In benchmarks using the one-million-row NYC Taxi dataset, chdb ran on four-vCPU, 32-GB ClickHouse Managed Postgres instances and showed the most consistent performance; pg\_duckdb and pg\_lake took about two to three times longer for CSV, JSON, and Parquet, while aws\_s3 supported fewer formats. Its chdb\_query() function executes chDB queries, while chdb\_hook integrates with COPY for AWS S3, Google Cloud Storage, Azure Blob Storage, local files, and HTTP URLs, with wildcard expansion for multiple files. The extension launches a separate helper application on demand and exchanges ClickHouse Native results through file descriptors, limiting resource use and isolating the PostgreSQL cluster.


### [Safety for Whom? Refusing the Right Subset of a Topic, Not the Whole Topic](https://yomu.fyi/post/safety-for-whom-refusing-the-right-subset-of-a-topic-not-the-whole-top.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Antonio Tiene, Alejo Lopez Avila, Iker García-Ferrero
- Published: Sep 8, 2026
- Type: Explainer

Topic-level safety systems can conflate a dangerous subject with dangerous intent, causing models to refuse legitimate requests alongside harmful ones. The paper studies deployment-specific boundaries using political persuasion as a testbed: refuse manipulation or targeted persuasion while answering factual political questions. Its training pipeline repairs self-generation coverage through escalating retries, adds verified surface-dangerous benign examples, and evaluates held-out harmful-benign prompt pairs rather than relying on aggregate refusal rates. On Qwen3-8B, escalated coverage raises in-distribution political refusal from 9.47% to 84.75%, while the strongest configuration reduces unsafe responses across three broader benchmarks from 26.26% to 0.14%; however, XSTest over-refusal rises to 74.00%. Adding benign boundary data reduces over-refusal on comply-worthy held-out prompts from 32.94% to 4.16%, with harmful-side refusal declining from 91.88% to 87.72%.


### [Optimize vLLM speculative decoding with FastMTP heads](https://yomu.fyi/post/optimize-vllm-speculative-decoding-with-fastmtp-heads.md)
- Company: [Red Hat](https://yomu.fyi/company/red-hat.md)
- Author: Rahul Tuli
- Published: Sep 8, 2026
- Type: Tutorial

Autoregressive LLM inference is memory-bandwidth bound because each token requires a full pass over billions of parameters, while speculative decoding can verify several drafted tokens in one pass. Speculators 0.6.0 adapts a single native MTP head with FastMTP-style recursive fine-tuning, addressing the mismatch between teacher-forced training and vLLM’s autoregressive reuse of the head. The training loop fuses each next-token embedding with the current hidden state, predicts the following token, feeds the resulting hidden state back, and weights three steps with normalized exponential decay using β = 0.6. Native mtp.\* weights extraction, shared full-vocabulary embeddings and output head, and vLLM-compatible stitching support deployment; on Qwen3-Next-80B-A3B-Instruct evaluated on GSM8K, mean accepted length improved at positions 0–2 from 0.897, 0.719, 0.476 to 0.912, 0.776, 0.616.


### [dotInsights | September 2026](https://yomu.fyi/post/dotinsights-september-2026.md)
- Company: [Jetbrains](https://yomu.fyi/company/jetbrains.md)
- Author: Rachel Appel
- Published: Sep 8, 2026
- Type: Announcement

dotInsights’ September 2026 issue curates recent .NET and software development material from the developer community, alongside JetBrains announcements and a short social-media coffee break. The featured links cover Azure Service Bus filters, .NET 11 performance, Fetch Metadata-based CSRF protection, software-architecture complexity, standalone hosting for the .NET Aspire Dashboard, and several C# and ASP.NET Core topics. Other selections address AI coding agents, local AI models, MCP clients in .NET, Blazor CSV import and full-text search, URI handling, digital signatures, memory usage in Kubernetes, and .NET MAUI. The issue also notes that Roslyn, the C# compiler, is open source, written in C#, and exposes APIs for analyzing and generating code, while JetBrains promotes upcoming events, Rider and ReSharper releases, and Rider’s AI-agent refactoring capabilities.


[Newer posts](https://yomu.fyi/posts/page/53.md) · [Older posts](https://yomu.fyi/posts/page/55.md)