Loading…
Browse by type
Problem & solution reads
Problem & solution posts on Yomu, summarised newest first.
- 916 posts
- 20 companies
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Problem & solution posts
Scaling developer experience: How we improved Android Studio in a large monorepo
Grab’s passenger Android monorepo contains roughly 2,000 modules and 11 million lines of code, making Gradle-based Android Studio syncs increasingly slow and memory-intensive even though Bazel handles builds. A cold full sync could exceed 35 minutes, and a survey of 55 developers found that 82% wanted an option to exclude modules from syncing. The team built a custom Focus plugin that uses a dependency graph generated during migrateToBazel, avoids repeating Gradle’s configuration phase, and integrates with Android Studio through a one-click Select → Focus → Sync flow; a Groovy include-shadow technique avoids requiring teams to reorganize settings files. Typical focused syncs fell below two minutes, while reported results show reductions from 35 minutes to under one minute depending on module size, memory usage from 10 GB to 2 GB, and improved Compose preview updates, with broader IDE operations requiring expanded focus or full sync.
From latency to instant: Modernizing GitHub Issues navigation performance
GitHub Issues addressed navigation latency by shifting workloads to the client using a local-first, stale-while-revalidate architecture. To evaluate perceived delays, the team tracked user transitions through Highest Priority Content thresholds, aiming for instant rendering in under 200 milliseconds. The core implementation added a persistent client-side cache using IndexedDB, paired with a synchronous in-memory tier to serve hot issue payloads without asynchronous overhead. A selective preheating mechanism resolves missing cache entries for high-intent links without overburdening backend capacity with redundant requests. Following broad rollout, the proportion of instant React soft navigations increased from 4% to approximately 22%, yielding an overall cache-hit ratio of around 33%.
amazon.science
Promptimus: Improving already good LLM prompts with zero manual engineering
Enterprise applications rely on mature large language model prompts that incorporate intricate business logic and regulatory constraints, making further optimization and cross-model migration challenging. Promptimus automates prompt refinement through a four-step iterative loop that evaluates candidate prompts against decomposed metric checkpoints, diagnoses failure bottlenecks, and generates targeted improvement strategies. The system operates either via full prompt rewrites in standard mode or through surgical, programmatic find-and-replace edits in edit mode to preserve structured taxonomies and schemas. Tested across multimodal classification and code generation tasks, the framework achieves performance gains with small development datasets of 20 to 50 samples. Amazon is making the model-agnostic optimization system available through Amazon Bedrock to support enterprise generative-AI model migration without manual engineering.
The Rosetta stone of CPS: Claroty’s AI-powered library
CPS asset identification is difficult because 88% of CPS assets do not transmit an exact product code and 76% use product codes that differ from the vendor's official records. Claroty's AI-Powered CPS Library addresses this entity-resolution problem with a hybrid architecture that combines classic matching, statistical inference, generative AI, NLP, reasoning agents, and human-in-the-loop review to reconcile protocol data, vendor records, manuals, and vulnerability information. Its Databricks-based lakehouse uses Delta Lake, Unity Catalog, Spark pipelines, Model Serving, MLflow evaluations, Lakebase, and Databricks Apps to maintain governed, auditable mappings across a catalog of more than 17 million assets. The post reports 25% improved vulnerability-attribution accuracy and says early tests gave 56% of analyzed devices new or updated recommendations for outdated firmware, while low-confidence mappings remain subject to expert review.
Clinical operations intelligence belongs on the Lakehouse
The Site Feasibility Workbench addresses fragmented clinical operations data by placing a site-selection decision-support application, its models, and data on Databricks. Released as a fully open-source Databricks App, it combines a FastAPI backend, React frontend, Unity Catalog-governed data, Lakebase operational state, and AI/BI Genie within the workspace. TA-segmented LightGBM models use the organization’s CTMS, EDC, and IRT history alongside real-world evidence, patient access, qualification, Open Payments, and protocol-execution signals to rank sites. Each prediction stores SHAP attributions in a governed Unity Catalog Delta table, while MLflow tracks training runs and model versions; saved shortlists persist to Lakebase and Genie answers governed-data questions. The six-step workflow supports protocol selection through final shortlisting, including diversity considerations; the release is a decision-support layer rather than a source of record, and deployment reportedly takes about 30 minutes before security review and validation.
How CFOs in consulting can recover margin with Databricks
Consulting firms are facing margin pressure despite growing deal pipelines, with 2024 EBITDA margins at 9.8%, utilization at 68.9%, and revenue growth at 4.6%. The post attributes this to fragmented finance workflows spanning Salesforce, SAP, Workday, Concur, and spreadsheets, plus inconsistent metrics and manual reconciliation. It proposes using Databricks to create a governed data layer with Lakeflow, Lakehouse, and Unity Catalog, then adding Genie for natural-language analysis and finance apps that trigger workflows and write approvals back in real time. A global consulting firm reportedly cut cash forecasting cycles by 3–5 days and reduced reporting-cycle FTE hours by 80%, while Databricks’ own close fell from 15 days to 8. The proposed rollout spans 10–12 weeks and begins with one tightly scoped workflow such as revenue close, unbilled WIP, T&E leakage, or practice-margin analysis.
An Engineer’s Guide to Better AI Skills: Implementing a Testing Process to Optimize Agent…
Software engineers deploying AI agents often encounter unreliable skill invocation for repository-specific architectures, such as Pinterest's iOS framework. To quantify and resolve this issue, a Bash test harness piped positive and negative prompts into Pin-agent and Claude Code while parsing streamed JSON logs for execution signatures. Initial vanilla benchmarks yielded baseline accuracies of 73% for GPT 5.2-codex and 62% for Opus 4.5 Claude. Optimizations such as expanding YAML frontmatter descriptions with architectural context, utilizing capitalized imperative directives, and adding reference tables in AGENTS.md raised invocation rates. Both agents reliably loaded skills when prompts contained explicit invocations, demonstrating that empirical testing paired with descriptive prompting stabilizes agent skill adoption.
How we achieved truly serverless GPUs
GPU inference workloads suffer from spiky demand, leading to low GPU Allocation Utilization when capacity is fixed. Modal reduced replica startup latency from over 2 kiloseconds to about 50 seconds using four techniques. A buffer of idle GPUs removes instance allocation from the hot path. A custom content-addressed filesystem serves container images lazily from a multi-tier cache. CPU and CUDA checkpoint/restore fast-forward through initialization by directly restoring processes and GPU contexts into memory. These optimizations enable truly serverless GPU inference, as demonstrated by Reducto’s document processing scaling to thousands of GPUs with cold starts reduced from ~70s to ~12s.
We replaced Redis with MySQL for inventory reservations—and it scaled
Shopify replaced its Redis-based inventory reservation system with MySQL to unify reservations with the inventory ledger and provide ACID coordination between payment and inventory updates. The new design stores one row per sellable unit in a bounded pool of up to 1,000 rows per item/location combination, uses MySQL 8's SKIP LOCKED to avoid waiting on locked units, and replenishes depleted pools. Composite primary keys reduced row locks, READ COMMITTED avoided gap locks that interfered with replenishment, and consistent lock ordering prevented deadlocks. The investigation found connection-pool usage elsewhere in checkout, rather than reservation query speed alone, was the immediate ceiling; cleanup removed 50% of primary-database reads and 33% of transactions, while configuration changes added headroom. Shopify validated the design in shadow mode, then rolled it out gradually with a Redis rollback path, meeting high-throughput targets during peak 2025 traffic.
Unlocking the Archives: Turning Unstructured Documents into a Searchable Database for Groundwater Discovery
MapAid and the Sudan Association for Archiving Knowledge (SUDAAK) needed to make nearly 700 scanned geological reports searchable and recover well data for MapAid’s WellMapr groundwater models. Databricks built a Databricks-only pipeline that renders pages as images, uses multimodal AI Functions for page classification and schema-constrained JSON, and applies targeted sampling to cut classification processing by more than 70%. Water-relevant documents receive page-by-page OCR through the Foundation Model API, with well and borehole identifiers linking information across pages before structured records capture locations, depths, static water levels, and yields. In its first full run, 654 documents and 5,570 pages were classified in under three hours; 95% received excellent or good evaluations, about half contained water data, and 299 well or borehole records were extracted.
Fighting Tool Sprawl: The Case for AI Tool Registries
Rapid enterprise AI agent adoption without centralized infrastructure introduces widespread tool sprawl, duplicated engineering effort, and severe operational opacity. Individual application teams frequently build ad hoc, undocumented integrations, leaving security teams unable to discover, audit, or approve active agent capabilities. Internal enterprise tool registries address this coordination failure by establishing an organization-specific catalog tailored to internal security policies, data, and compliance requirements. A production-grade registry delivers core discovery and version tracking functionality while surfacing security certification metadata and enabling consistent identity-scoped access control across teams. Organizations establishing centralized tool infrastructure can accelerate agent onboarding, implement deny-by-default governance policies, and mitigate technical debt before uncoordinated deployments compound across the enterprise.
Retail markdown optimization: from reactive markdowns to proactive
Retail markdown optimization addresses the gap between changing demand signals and delayed merchandising action. The post defines it as strategically reducing prices on slow-moving or end-of-life inventory, using demand forecasts, sell-through rates, weeks of supply, and price-elasticity models to choose which SKUs to mark down, when to start, how deep to discount, and where to act. It presents Databricks Genie as a natural-language interface across e-commerce, store, and wholesale data, with supplier, margin, and historical-pattern context; one example is Coop’s Microsoft Teams assistant, AskCap, which reported a 30% retention rate among internal users. Earlier detection of sell-through deceleration can give merchants more time to protect margin, adjust open-to-buy, and reallocate capital, while Genie supports decisions rather than making them.
From vibe code to production-ready: observability for Next.js and Supabase apps
AI-assisted software development often generates functional code that lacks proper instrumentation, leading to unindexed database queries, N+1 patterns, and fragmented connection strategies. While Supabase provides built-in query analytics, security advisors, and subsystem logs, it cannot trace distributed requests across the entire application stack. Integrating Sentry bridges this visibility gap by establishing distributed traces across Next.js frontends, Deno-based Supabase Edge Functions, and Postgres databases. Teams can forward Supabase logs into dedicated Sentry projects and configure runtime integrations to automatically detect performance bottlenecks such as slow spans and Web Vitals regressions. Furthermore, integrating Model Context Protocol servers allows Sentry's Seer debugger to identify root causes and generate automated fixes for emerging issues.
Powering self-driving vehicle analytics at Avride with ClickHouse Cloud
Avride’s expanding autonomous passenger vehicle and delivery robot fleet generated thousands of data points per minute, overwhelming its legacy Apache Iceberg and Parquet metadata indexing architecture. Iceberg’s optimistic concurrency model struggled with parallel writes, pushing ingestion latencies to hours or days, inflating query times to 10–20 seconds, and requiring raw data duplication across storage locations. To resolve these bottlenecks, Avride built a custom indexing and metrics warehouse powered by ClickHouse Cloud on AWS backed by Amazon S3 object storage. The new architecture dropped ingestion delays to seconds, reduced worst-case lookup latencies to under one second, eliminated the need for bespoke caching layers, and removed payload duplication. Beyond operational analytics, the unified ClickHouse platform now supports internal ride-replay debugging tools, simulation workflows, and high-cardinality vehicle C++ performance trace profiling.
Upgrading M3DB from v1.1 to v1.5
Uber upgraded M3DB from v1.1 to v1.5 after three years on its previous open-source deployment, seeking a repeatable rollout process for a roughly 100-cluster, 5,500-node environment. The effort addressed scale-specific validation: offline benchmarks measured correctness, latency, success rates, bootstrap and tick durations, and persistence across restarts and replacements, while online shadow testing compared sampled production traffic between old and new versions. Within each cluster, upgrades were limited to one node at a time because writes used a two-thirds majority quorum. A rollout-related read-quorum incident exposed missing replica data; the cause was bootstrap misconfiguration, corrected by using “exclude_commitlog,” since production workloads did not employ WAL. Production then showed significantly lower memory usage and improved node bootstrap times, supporting continued, data-driven upgrades and contributions to M3DB's benchmarking and documentation.
How Superhuman and Databricks built a 200K QPS inference platform together
Superhuman replaced a DIY vLLM-based serving stack for its grammar-correction model with Databricks model serving, targeting peak traffic above 200,000 QPS, sub-second P99 latency, four-nines reliability, and no quality regression. The platform combines an Endpoint Discovery Service with power-of-two-choices load balancing, concurrency-based autoscaling, and asymmetric scale policies to reduce hotspots and handle rapid traffic ramps. Lazy-loading container images changed startup from several minutes to a few seconds by fetching metadata first and retrieving 4MB blocks on demand. Runtime work on H100 GPUs raised per-pod throughput from 750 to 1,200 QPS: FP8 weight quantization contributed up to 30%, while a multiprocessing server added 20%, alongside C++ and asynchronous scheduling optimizations. The teams validated changes through shadow testing and evaluation harnesses, and Superhuman is migrating additional models and workflows to the platform.
Enhancing Ad Relevance: Integrating Real-Time Context into Sequential Recommender Models
Pinterest's sequential ads candidate generator inferred user embeddings offline from historical offsite conversions, which lacked real-time browsing context. Consequently, less than 1% of impressions on the Related Pins surface were attributed to this generator because its retrieved candidates struggled in downstream ranking. To resolve this limitation, engineers developed the Contextual Sequential Two Tower Model, integrating a context layer directly into the query tower alongside user demographic features. Offline training utilized synthetic context derived from positive conversion labels paired with high dropout rates to preserve sequence reliance, while serving split into offline Transformer inference and online context computation. Offline evaluations showed a 3x to 10x increase in Recall@K, and production deployment lifted median candidate relevance by 275% to 300% and overall ROAS by approximately 0.7%.
Addressing HR's widening capacity gap with AI
HR teams face a widening capacity gap as strategic expectations, complex employee issues, workforce volatility, skills shortages, and demands for personalized support collide with largely unchanged headcount and tools. The article presents AI transformation as an incremental journey: first establish a secure Employee 360 from structured and unstructured enterprise data, then build reusable workforce insights, augment workflows with human oversight, and progress toward broader transformation. It emphasizes data governance, including access controls, auditing, quality, standardization, and reliable interpretations, while noting that trust has limited AI’s business impact so far. MathCo and Databricks support this roadmap through NucliOS, whose Data Studio, AI Studio, and Decision Studio environments connect governed data, explainable models, feedback loops, and decision applications; Databricks supplies the lakehouse foundation, lineage, quality checks, and privacy-compliant access.
How we turned support into a revenue engine at Intercom
After Fin improved resolution rates and efficiency at Intercom, support gained capacity and began testing a consultative role focused on customer value and company goals. The team generated campaign ideas through a support town hall, then assigned a small volunteer group to proactively help self-serve customers with Fin trials, optimization, and adoption. Volunteers were trained to probe customer context rather than only resolve immediate issues, while sales and success teams provided guidance on timing and outreach. Over six months, engaged accounts grew roughly twice as fast in usage and expansion as accounts contacted without a response; the team also tracked resolutions, automation, and influenced expansion ARR. The initiative expanded into cross-functional partnerships and a ~16-person volunteer group, positioning support as a contributor to adoption, retention, and expansion revenue.
Energy trading analytics in a real-time market
Energy trading markets settle every 15 minutes, while many analytics environments still rely on nightly batch pipelines, creating latency between changing conditions and trading decisions. The post presents Databricks Genie as a conversational interface that lets traders and portfolio managers query actual market data, weather integrations, historical position data, and current books. Its example combines a node’s 90-day natural-gas basis distribution with a Northeast temperature anomaly, while the described system can query real-time and historical data together. Genie also includes settlements, curtailments, and ancillary-service data, with governed access so users see only authorized information. The conclusion is that faster data access can remove information latency and compound margin without replacing trading expertise; the source states that Genie is available today.
Top topics
- Architecture834
- Performance816
- LLMs660
- Machine Learning622
- Open Source563
- AI Agents553
- AI470
- Developer Experience448
Weekly digest
The best engineering reads, once a week.