---
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

### [Get your VLM running in 3 simple steps on Intel CPUs](https://yomu.fyi/post/get-your-vlm-running-in-3-simple-steps-on-intel-cpus.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Ezequiel Lanza, Helena, Nikita, Ella Charlaix, Ilyas Moutawwakil
- Published: Oct 15, 2025
- Type: Tutorial

Deploying Vision Language Models locally presents computational challenges despite providing data privacy and independence from external servers. Optimum Intel and OpenVINO address this by optimizing lightweight models like SmolVLM2-256M for low-resource hardware, including Intel CPUs and GPUs. The process involves exporting the model to the OpenVINO Intermediate Representation format, applying 8-bit weight-only or static quantization, and executing inference with the transformed model. In single-image CPU benchmarks on an Intel Core Ultra 7 265K processor, OpenVINO with 8-bit weight-only quantization reduced time to first token from 5.150 seconds under PyTorch down to 0.247 seconds. The optimized configuration also raised decoding throughput from 0.722 to 63.928 tokens per second.


### [Nemotron-Personas-India: Synthesized Data for Sovereign AI](https://yomu.fyi/post/nemotron-personas-india-synthesized-data-for-sovereign-ai.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Kiran Praveen, Utkarsh Vaidya, Evan A, Lipika Ramaswamy, Dhruv Nathawani, Dane Corneil, Yev Meyer
- Published: Oct 13, 2025
- Type: Announcement

Most open artificial intelligence datasets reflect Western norms and English-only contexts, limiting model effectiveness across India's multilingual and multi-script landscape. To address this shortage of culturally grounded training data, NVIDIA released Nemotron-Personas-India under a CC BY 4.0 license. The synthetic dataset contains 21 million personas across 3 million records spanning 7.7 billion tokens in English, Devanagari Hindi, and Latin-script Hindi. The generation pipeline utilized NeMo Data Designer, an Apache-2.0 licensed Probabilistic Graphical Model for demographic grounding against 2011 Census distributions, and GPT-OSS-120B for narrative text. All records are entirely synthetic to eliminate re-identification risks while enabling fine-tuning for region-aware AI agents and domain-specific copilots.


### [Understanding Etsy’s Vast Inventory with LLMs](https://yomu.fyi/post/understanding-etsy-s-vast-inventory-with-llms.md)
- Company: [Etsy](https://yomu.fyi/company/etsy.md)
- Author: Vipul Setty
- Published: Oct 13, 2025
- Type: Problem & solution

Etsy’s seller-created inventory spans more than 100 million items and thousands of categories, making it difficult to capture attributes such as color, material, or niche properties without global SKUs. Because sellers often provide descriptions, titles, and photos instead of complete structured fields, Etsy uses LLMs to distill unstructured listing data into attributes for search filters, comparisons, and other experiences. The pipeline combines seller data, expert-selected few-shot examples, taxonomy logic, and category rules; parallel prompts are routed through LiteLLM, then responses are validated with Pydantic dataclasses and monitored with tracing, metrics, alerts, and performance checks. Human-reviewed silver labels support evaluation, while target categories increased complete attribute coverage from 31% to 91%, and inferred attributes improved search-filter engagement and overall post-click conversion.


### [Cars24 Improves Search For 300 Million Users With MongoDB Atlas](https://yomu.fyi/post/cars24-improves-search-for-300-million-users-with-mongodb-atlas.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Author: Nick Bell
- Published: Oct 12, 2025
- Type: Problem & solution

Online car marketplace Cars24 previously operated a dual architecture where relational databases like Postgres synchronized data into separate search engines like Elasticsearch. As service offerings expanded globally for 300 million users, multiple engineering teams piping data into a single search index caused severe synchronization challenges, operational overhead, and developer friction. To simplify this infrastructure, Cars24 migrated to MongoDB Atlas hosted on Google Cloud, utilizing integrated Apache Lucene-powered MongoDB Atlas Search directly within the database layer. This architectural shift eliminated synchronization pipelines while accelerating search feature delivery through a unified API. Cars24 also replaced its ArangoDB geospatial search deployment with MongoDB Atlas, achieving robust multi-document ACID transactions across shards and cutting operational costs by 50%.


### [Arm will be @ PyTorch Conference, Join Us!](https://yomu.fyi/post/arm-will-be-pytorch-conference-join-us.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: EricSondhi
- Published: Oct 10, 2025
- Type: Announcement

Arm announced its schedule and offerings for the PyTorch Conference on October 22–23, focusing on building and deploying AI applications using PyTorch and ExecuTorch. The company is hosting one-on-one workshops with design experts to improve product usability and introduce Yellow Teaming practices for mitigating fairness, transparency, privacy, and security risks. Additionally, Arm is running 30-minute Voice of the Developer sessions to gather feedback on real-world challenges, including migrating from NVIDIA or x86 to Arm, debugging models at scale, and running large language models on edge devices. At Booth P1, attendees can view interactive demonstrations featuring neural graphics training, speech recognition, audio generation, vLLM, Mixture of Experts, and ExecuTorch edge deployments.


### [The Cost of Not Knowing MongoDB, Part 3: appV6R0 to appV6R4](https://yomu.fyi/post/the-cost-of-not-knowing-mongodb-part-3-appv6r0-to-appv6r4.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Author: Artur Costa
- Published: Oct 9, 2025
- Type: Benchmark

To alleviate a disk throughput bottleneck on a MongoDB server, this optimization pass focuses on shrinking document sizes across application revisions appV6R0 through appV6R4. The implementation replaces an array of items with a dynamic schema document, using date segments such as days or combined months and days as field names to store status totals. Redundant date components already encoded in the binary \_id field are omitted from internal document keys. Bulk upsert operations rely on updateOne with the $inc operator to update matching date keys or create documents when missing. Aggregation pipelines process reports using the $objectToArray and $reduce operations to calculate totals across date intervals while preserving a single index on \_id.


### [How AI Agents are Evolving Shopify's Product Taxonomy at Scale (2025) - Shopify](https://yomu.fyi/post/how-ai-agents-are-evolving-shopify-s-product-taxonomy-at-scale-2025-sh.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: 2023-10-18
- Published: Oct 9, 2025
- Type: Explainer

Shopify’s product classification system processes over tens of millions of predictions daily, while its taxonomy contains more than 10,000 categories and 2,000 attributes that must adapt as commerce changes. Manual curation created bottlenecks, expertise gaps, and inconsistencies that could reduce discoverability, search quality, filtering, and classification quality. The proposed AI multi-agent system combines structural analysis with product-driven analysis grounded in merchant titles, descriptions, and categories, then uses intelligent synthesis, equivalence detection, and quality assurance to recommend taxonomy improvements. Agents can search and validate hierarchical relationships, while human experts retain oversight; Shopify tested the method in Electronics > Communications > Telephony against its previous manual expansion approach and presents it as a proactive, scalable way to improve taxonomy consistency before gaps affect merchants and customers.


### [Unlocking Faster Insights with Experimenter-Defined Segmentations](https://yomu.fyi/post/unlocking-faster-insights-with-experimenter-defined-segmentations.md)
- Company: [Etsy](https://yomu.fyi/company/etsy.md)
- Author: Athena Chang
- Published: Oct 8, 2025
- Type: Problem & solution

At Etsy, Catapult supported only 13 hard-coded segmentations, forcing experimenters who needed other cuts to rely on slow, ad-hoc SQL analysis outside the platform. They replaced that fixed model with a Segmentation Library and creation tool where experimenters define reusable segmentations as SQL queries plus join columns connecting segment data to experimental units. Each definition is validated before production, then processed through a configuration-driven pipeline that detects new segmentations and generates independent tasks, improving failure isolation and allowing downstream work to wait only for required results. In the first three months, teams added 34 segmentations, increasing support from 13 to 47, while 78% of experimenters reported saving 2–5 hours monthly through custom analysis in Catapult.


### [Kurt Got Got](https://yomu.fyi/post/kurt-got-got.md)
- Company: [Fly.io](https://yomu.fyi/company/fly-io.md)
- Published: Oct 8, 2025
- Type: Incident / postmortem

Fly.io recounts the takeover of its X/Twitter account after CEO Kurt Mackey entered credentials on members-x.com in response to a phishing email from alerts-x.com. The attackers revoked existing tokens, configured new 2FA, changed the account email, and used it for a crypto scam; the company says no customer systems or accounts were at risk, though the incident caused more than 15 hours of brand damage. Fly.io attributes the exposure to treating the shared social account as a legacy exception rather than protecting it through its normal Google IdP and phishing-resistant MFA. The post explains how FIDO2, U2F, and Passkeys prevent fake sites from receiving real credentials, while noting that the 1Password browser plugin could also have flagged the mismatched domain. The account is now protected with Passkeys.


### [BigCodeArena: Judging code generations end to end with code executions](https://yomu.fyi/post/bigcodearena-judging-code-generations-end-to-end-with-code-executions.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Terry Yue Zhuo
- Published: Oct 7, 2025
- Type: Announcement

Assessing AI-generated code via static inspection is error-prone and fails to reflect whether code functions correctly across real-world environments. BigCodeArena addresses this evaluation bottleneck as a human-in-the-loop platform that executes model outputs in real time inside isolated sandbox environments across ten programming languages and eight execution frameworks. Users interact directly with live applications, refine implementations through multi-turn conversations, and submit pairwise preference votes to generate Elo ratings via the Bradley-Terry model. In five months of community testing covering over 14,000 conversations, proprietary models including o3-mini and o1-mini led overall Elo rankings, while open models like Qwen2.5 and Llama-3.3-70B trailed closed competitors. The project has open-sourced its full evaluation pipelines, conversation datasets, and related benchmarks to support community-driven code generation assessment.


### [Medium Android App — Migrating from Apollo Kotlin 3 to 4: Lessons Learned](https://yomu.fyi/post/medium-android-app-migrating-from-apollo-kotlin-3-to-4-lessons-learned.md)
- Company: [Medium](https://yomu.fyi/company/medium.md)
- Author: Pierrick CAEN
- Published: Oct 6, 2025
- Type: Tutorial

Migrating the Medium Android app from Apollo Kotlin 3 to 4 exposed cache and exception-handling issues that were hidden by the earlier implementation. Apollo Kotlin 4 uses a new group, plugin, and package identifier, places exceptions in ApolloResponse, and replaces ApolloCompositeException with a primary exception plus suppressed failures. CacheMissException errors appeared in watch() because programmatic cache IDs were incomplete or incorrect; the app addressed this with declarative type and field policies, \_\_typename on every operation, cache-update helpers, and policy-aware safeWatch filtering. The migration also replaced executeV3(), removed fetchThrows = true, added a Currency scalar adapter, and identified HTTP batching and persisted queries as future improvements, while concluding that Apollo 4 produced a more robust cache integration after the underlying issues were fixed.


### [Highly concurrent in-memory counter in GoLang](https://yomu.fyi/post/highly-concurrent-in-memory-counter-in-golang.md)
- Company: [Grab](https://yomu.fyi/company/grab.md)
- Author: Naveen Kumar Jakuva Premkumar
- Published: Oct 6, 2025
- Type: Problem & solution

An overloaded database faced heavy traffic from campaign usage tracking, motivating an in-memory counter rather than an immediate SQL-to-NoSQL migration. Because exact counts were not required, increments could be cached in the application server and flushed periodically to reduce database queries and CPU use. The implementation uses Go’s Sync.Map, with LoadOrStore and CompareAndSwap for concurrent increments, plus LoadAndDelete to atomically remove values for asynchronous persistence; the source notes a fresh-map swap as an alternative locking optimization. On an Apple M4 Pro, the 2,000-key benchmark reported 53 ns/op and 1 billion operations per 60 seconds, versus 160 ns/op and 466 million operations for a Mutex-protected map, but Sync.Map used 88 B/op and four allocations. In production, update queries fell from 140 QPS to 45 QPS, master CPU from 35% to 18%, and replica CPU from 19% to 12%.


### [Innovating with MongoDB | Customer Successes, October 2025](https://yomu.fyi/post/innovating-with-mongodb-customer-successes-october-2025.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Author: Katie Palmer
- Published: Oct 2, 2025
- Type: Announcement

Enterprise organizations are modernizing legacy monoliths, mainframes, and relational databases to improve scalability and system performance. MongoDB launched its Application Modernization Platform, an AI-powered program combining artificial intelligence tooling with migration methodologies. Using these capabilities, Wells Fargo implemented an operational data store to serve 40 percent of external vendor traffic with sub-second response times across more than 7 million transactions. CSX utilized MongoDB Atlas on Azure alongside Cluster-to-Cluster Sync to complete a cloud migration within hours while sustaining continuous operations. Additionally, Intellect Design reduced onboarding workflow times by 85 percent, and Deutsche Telekom scaled daily customer interactions from under 50,000 to approximately 1.5 million.


### [From workspace to runway: How Kim Shui used Notion for New York Fashion Week](https://yomu.fyi/post/from-workspace-to-runway-how-kim-shui-used-notion-for-new-york-fashion.md)
- Company: [Notion](https://yomu.fyi/company/notion.md)
- Author: Niki Frias
- Published: Oct 2, 2025
- Type: Explainer

Kim Shui’s Spring 2025 New York Fashion Week show took place on the 102nd floor of One World Trade Center, requiring months of planning for a 10-minute presentation. Organization was central because the event involved guest logistics, multiple teams, and coordination in an unusual venue. Shui used Notion as a single hub for schedules, mood boards, run-of-show materials, to-dos, budgets, and real-time team updates. She also collected sketches, references, textures, notes, and fabric swatches there, using digital documentation to connect creative ideas, communicate direction, and build a searchable, cross-referenced archive. She says mapping timelines and deliverables beside creative boards helps account for production constraints while preserving room for creative exploration.


### [The 10 Skills I Was Missing as a MongoDB User](https://yomu.fyi/post/the-10-skills-i-was-missing-as-a-mongodb-user.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Author: Joel Lord
- Published: Oct 2, 2025
- Type: Explainer

Transitioning from relational databases to MongoDB often leads developers to replicate SQL habits, causing severe performance bottlenecks, brittle schemas, and operational challenges. Effective document design requires balancing referencing and embedding based on query access patterns and update frequencies, rather than placing every entity into separate collections or excessively nesting large documents. Database performance depends on matching indexes to specific query shapes and field orders, which developers can systematically verify by inspecting execution plans using explain plans. Utilizing the built-in aggregation framework allows filtering, grouping, and calculations to run natively inside the database, eliminating slow, manual data transformations in application code. Establishing early monitoring for latency, replication lag, memory usage, and cluster failover resilience ensures stability before adopting advanced tools like Atlas Search and Vector Search.


### [Litestream v0.5.0 is Here](https://yomu.fyi/post/litestream-v0-5-0-is-here.md)
- Company: [Fly.io](https://yomu.fyi/company/fly-io.md)
- Published: Oct 2, 2025
- Type: Announcement

Litestream v0.5.0 updates the SQLite backup and restore tool with faster operation and efficient point-in-time recovery (PITR). It replaces WAL segment generations with LTX files, an interchange format for ordered page ranges that supports compaction and monotonically incrementing transaction IDs. Hierarchical compaction combines changes across 30-second, five-minute, and hourly levels, allowing restoration to any point in time using only about a dozen files on average and performance limited by I/O throughput. The release keeps old v0.3.x WAL files intact and preserves configuration compatibility, but permits only one replica destination per database and renames litestream wal to litestream ltx. It also adds per-page compression and indexing, removes CGO through modernc.org/sqlite, supports NATS JetStream replicas, updates cloud-storage clients, and previews a VFS for read replicas that can fetch pages from S3 while hydration continues.


### [SOTA OCR with Core ML and dots.ocr](https://yomu.fyi/post/sota-ocr-with-core-ml-and-dots-ocr.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Christopher Fleetwood, Pedro Cuenca
- Published: Oct 2, 2025
- Type: Tutorial

Running competitive models like RedNote's 3B parameter dots.ocr on-device provides zero-cost, network-free optical character recognition while leveraging power-efficient hardware like Apple's Neural Engine. Because the Neural Engine requires Core ML, converting the 1.2B parameter NaViT vision encoder from PyTorch requires resolving multiple graph tracing incompatibilities. The initial conversion pipeline targets FLOAT32 execution on GPU by capturing the execution graph with torch.jit.trace and compiling it via coremltools. Simplifying the model for single-image inference eliminates dynamic sequence masking, dynamic tensor iteration, and multi-attention complexity. Although the resulting Core ML model accurately matches PyTorch baseline numerical precision, the unoptimized artifact spans over 5GB and takes over one second per forward pass.


### [Smarter AI Search, Powered by MongoDB Atlas and Pureinsights](https://yomu.fyi/post/smarter-ai-search-powered-by-mongodb-atlas-and-pureinsights.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Author: Kamran Khan (CEO, Pureinsights), Prasad Pashte
- Published: Oct 1, 2025
- Type: Announcement

MongoDB Atlas has announced the general availability of its integration with the Pureinsights Discovery Platform to deliver a unified keyword, vector, and generative search experience. The combined solution pairs MongoDB Atlas Search for standard text matching with MongoDB Atlas Vector Search and Voyage AI embeddings to interpret ambiguous or multilingual queries. Pureinsights provides the orchestration layer that ingests content, coordinates retrieval, and integrates large language models such as GPT-4 to generate cited responses. Users can customize generated outputs according to preferred technical depth, length, language, and role-specific requirements. The architecture enables enterprises to apply retrieval-augmented generation and semantic search across technical documentation, community forums, and internal knowledge bases.


### [Introducing RTEB: A New Standard for Retrieval Evaluation](https://yomu.fyi/post/introducing-rteb-a-new-standard-for-retrieval-evaluation.md)
- Company: [Hugging Face](https://yomu.fyi/company/hugging-face.md)
- Author: Frank Liu, Kenneth Enevoldsen, Solomatin Roman, Isaac Chung, Tom Aarsen, Fődi, Zoltán
- Published: Oct 1, 2025
- Type: Announcement

Public embedding benchmarks often fail to measure true generalization because training sets overlap with public evaluation data, rewarding memorization over general search quality. The Retrieval Embedding Benchmark (RTEB) beta introduces a standardized evaluation framework designed for real-world enterprise retrieval applications. To prevent overfitting while preserving transparency, RTEB combines fully public datasets with private evaluation sets managed exclusively by MTEB maintainers. The suite spans 20 languages across domains like healthcare, law, finance, and code, measuring ranked retrieval quality using NDCG@10 on datasets of at least 1,000 documents. While currently limited to text-only retrieval and containing repurposed question-answering datasets, the benchmark provides clear signals when models degrade on unseen data.


### [Top Considerations When Choosing a Hybrid Search Solution](https://yomu.fyi/post/top-considerations-when-choosing-a-hybrid-search-solution.md)
- Company: [MongoDB](https://yomu.fyi/company/mongodb.md)
- Author: Jose Parra
- Published: Sep 30, 2025
- Type: Explainer

Natural language queries and generative AI workflows like retrieval-augmented generation have exposed the limitations of standalone vector search, particularly for tokens outside an embedding model's training corpus. Hybrid search addresses these gaps by merging traditional lexical precision with dense vector context. Platforms combine these modalities using techniques such as reciprocal rank fusion and relative score fusion, often complemented by downstream cross-encoders or learning-to-rank models. Architecture choices center on the trade-offs between lexical-first systems using inverted indexes and vector-first platforms employing sparse vectors. MongoDB has integrated vector search indexes alongside its existing lexical indexes, releasing native hybrid search functions across Atlas, Community Edition, and Enterprise Server to unify operational data and search retrieval.


[Newer posts](https://yomu.fyi/posts/page/168.md) · [Older posts](https://yomu.fyi/posts/page/170.md)