Loading…
Browse by type
Problem & solution reads
Problem & solution posts on Yomu, summarised newest first.
- 910 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
Tevogen Bio’s Journey to Streamlining Life
Tevogen Bio faced drug development costing upwards of $3 billion and requiring 10–12 years; selecting targets for its single-HLA-restricted SARS-COV2 product took 18–24 months of wet-lab work. Tevogen.AI partnered with Microsoft and Databricks to build and curate a multi-terabyte protein-sequence library and parallelize processing for machine-learning training. Databricks Data + AI Platform, Medallion Architecture, Unity Catalog, and distributed computing cut processing from 50 days to 24 hours under governed, fine-grained access controls. The dataset reached 24 million proteins, 16 billion datapoints, ~700 million unique peptides, and ~37 million cross-matched expert articles. An MLOps framework automated training, inference, monitoring, and retention; the alpha PredicTcell model used XGBoost and ESM models and reported 93–97% recall and 38–43% accuracy, while beta PredicTcell and alpha AdapTcell work began.
Fizz accelerates ecommerce analytics with Databricks SQL
Fizz.hu, an ecommerce marketplace backed by OTP Group, outgrew a Microsoft SQL Server and Power BI setup that relied on daily batch loads and separate virtual-machine Python workloads. The company migrated about 50 tables and several stored procedures to Databricks SQL, recreated core views, and moved to a lakehouse architecture using an MVP-first approach rather than redesigning everything. Completed in three months, the migration switched off SQL Server while keeping Power BI reports running on the new engine. Nightly processing fell from three to four hours to roughly 90 minutes, reports became consistently available by 4:30 a.m., Power BI refreshes dropped about 50%, and gigabyte-scale exports finish in minutes. Databricks now unifies SQL, Python, machine learning, governance work, and Hungarian-language Genie self-service, with about 20 users reclaiming roughly 20% of an analyst’s time.
Zero downtime patching lakebase part-1 prewarming
Lakebase addresses the workload disruption caused by planned PostgreSQL maintenance, where a restart clears in-memory buffer and local file caches. Although the database can return in about one second at P99, the source reports a roughly 70% pgbench TPS reduction while caches warm, potentially causing timeouts. Lakebase uses stateless Postgres compute replicas and disaggregated shared storage to prewarm a replacement compute before a scheduled restart. The control plane transfers cached-page information, loads pages from shared storage, and follows relevant WAL from Safekeepers, while the current primary continues serving traffic. After prewarming, Lakebase promotes the new compute with PostgreSQL’s pg_promote without restarting the server; tests on 10 GB pgbench workloads found nearly instant throughput recovery, especially for read-only queries.
Database Branching in Postgres: Git
Shared staging databases often drift as schemas diverge, test data accumulates, and developers avoid creating isolated environments because traditional pg_dump copies are slow and expensive. Databricks Lakebase Postgres addresses this with database branching, which creates fully isolated environments from a parent’s exact schema and data at a point in time. Branches use copy-on-write: they share underlying versioned storage until a branch changes, then record only new data, while each branch has independent compute and a separate connection string. Idle branches scale compute to zero, and storage cost tracks changes rather than the full database size. This supports per-developer, per-pull-request, per-test-run, and recovery workflows, including point-in-time branching within the restore window, with the source describing branch creation through the console, CLI, or API in under a minute.
How a global fintech scaled coding agent traffic with Dedicated Model Inference
A global fintech runs its coding assistant on GLM 5.2 through Together’s Dedicated Model Inference (DMI), but spiky engineering-hours traffic outpaced static capacity planning. DMI gives the customer’s teams direct control over endpoint provisioning, scaling, configuration, model changes, and performance data through APIs, UI, and CLI. The workload is peak-load and relatively low-TPS, with ISL p50/p90/p95 of roughly 81K/163K/178K tokens and RPS p50/p90/p95 of 3/6/7, so concurrency headroom mattered more than raw throughput; the deployment used dozens of B200s, multiple replicas, and 256K context. When a reshaped endpoint later reached near-100% prefill capacity, requests queued for one to three minutes and decode throughput fell to roughly 5 tokens per second, the team restored cache-session-aware routing and widened max-inflight-per-worker, pushing the change same-day with zero downtime.
How MakeMyTrip achieved millisecond personalization at scale with Databricks
MakeMyTrip needed sub-second personalization for “last-searched” hotels across its consumer and corporate travel businesses, serving millions of daily users. Spark Structured Streaming’s micro-batch mode consistently delivered one to two seconds, while adopting Flink would have added a second engine, duplicated logic and increased operational overhead. The team instead adopted Databricks Real-Time Mode (RTM), using continuous data flow, concurrent pipeline scheduling and streaming shuffle to process events with millisecond-level latency on Spark. Its unified pipeline merges B2C and B2B Kafka streams, enriches events with each user’s “Last N” hotels from Aerospike, and writes results to Redis for serving in under 50ms. After tuning partitions, backpressure and rate limits, the production deployment reduced P50 latency from approximately 1.23 seconds to 44ms, P99 latency from over one minute to about 500ms, and increased click-through rate by 7%.
How FSIs eliminate silos between clients, operations, and finance
Financial institutions often split portfolio, operations, actuarial, and finance work across separate systems, even when teams depend on the same portfolios, positions, liabilities, and transactions. Databricks presents a unified, governed platform combining Databricks SQL, Unity Catalog Metric Views, Lakebase, Genie, Databricks Apps, and Lakeflow Designer to connect analysis with operational action. A worked workflow follows an actuary's duration analysis, a portfolio manager's scenario modeling and mandate changes, operations' IBOR/ABOR reconciliation, and finance's approval of adjustments and ledger entries. Governance features including row-level security, column masking, RBAC, ABAC, and shared business semantics keep access and terminology consistent, while Lakebase records corrections in auditable, timestamped tables. The result described is a closed-loop process in which strategy, risk, and accounting use the same governed data rather than reconciling separate systems.
How Equinix cut operational overhead with a shared services architecture on Amazon EKS
Equinix replaced its self-managed Kubernetes environment on Amazon EKS to address operational sprawl caused by independently managed clusters and inconsistent infrastructure practices. The new shared services architecture assigns platform ownership to a cloud operations team while application teams focus on business services. It uses separate Workloads, Platform, and Network accounts: workloads run across two Availability Zones with ALB, Gateway API, Cilium, and Hubble, while shared data services, GitHub Runners, and platform services reside in the Platform account. The Network account provides cross-account and multi-Region connectivity through AWS Transit Gateway, Route 53 Resolver, and Direct Connect. Equinix reports 40% lower operational overhead, 4x higher deployment frequency, and adoption of a unified architecture across multiple business organizations.
The Web Search Your Agent Inherited Isn't Good Enough
An agent that keeps a company’s market view current must combine Databricks-hosted internal account records with external web signals such as funding rounds, leadership changes, product launches, and hiring surges. The initial implementation duplicates enrichment logic across Claude Code, Codex, and a direct model API, with inconsistent search results and no shared cost, governance, or audit trail. Omnigent consolidates the agent definition, model, tools, policies, limits, and model-call governance above those harnesses, while letting the engineer select a common search provider. Nimble’s Search API supplies live web data, while its Web Search Agents orchestrate multi-source retrieval and extraction, cross-check sources, return citations, and adapt to difficult sites and recurring retrieval paths. In Nimble’s testing, adding its search increased LLM benchmark accuracy from 46% to 71% and halved web search costs; the proposed Databricks setup then connects internal records and external context under one governance surface.
The guest journey, updated in real time: extending Airbnb’s sequence recommender with Chronon
Airbnb’s guest sequence recommender previously refreshed Transformer-based embeddings in a daily batch, leaving activity from earlier the same day unavailable until the following run and creating up to nearly two days of staleness. To close this gap, Airbnb added Push Mode, which emits notifications when streaming feature values commit, and Near-real-time Model Transform, which invokes the deployed sequence encoder inside that streaming pipeline. Each page view or search is merged with the guest’s long-term and short-term sequence data, triggers inference, and writes a fresh embedding to the same low-latency store used by batch serving. The resulting system typically updates embeddings in under a minute, often within 10–30 seconds, improved offline NDCG by 1.67%, and increased uncancelled bookings by approximately one-third of a percent in online A/B tests.
How Real-Time Rewards Drive 2.5x Higher Customer Spend
Many retail loyalty programs use batch processing, delaying rewards by up to 72 hours or 10 business days, eroding customer trust. Nearly 75% of shoppers prefer instant rewards, 82% of Gen Z and Millennials expect real-time responsiveness, and real-time credit can increase basket size by 36% and spending by 2.5 times. Relational databases struggle with atomic writes across multiple tables, forcing batch cycles; layering microservices on top only distributes fragmented data without solving latency. MongoDB’s document model solves this by storing the complete loyalty profile in a single document for atomic writes; Triggers and Change Streams push updated balances instantly and allow promotional rule changes without redeployment. This architecture turns the post-purchase moment into immediate reinforcement, reduces support inquiries, and drives engagement, as seen with Starbucks Rewards contributing nearly 60% of total revenue.
DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models
DACA-GRPO addresses two weaknesses in reinforcement learning for diffusion language models: existing methods do not assign credit across denoising steps, and mean-field likelihood estimates are biased and high-variance. It is a lightweight, plug-and-play enhancement for GRPO-style trainers that combines Denoising Progress Scores with Stratified Masking Likelihood. The progress scores derive per-token importance weights from intermediate predictions without an additional forward pass, while stratified masking partitions token positions so each token is predicted with most of the sequence as context. Applied to three GRPO base methods, the approach improves results across seven benchmarks covering mathematical reasoning, code generation, constraint satisfaction, and constrained generation. Reported gains reach 5.6 percentage points on math reasoning, 7.4 on code generation, 36.3 on constraint satisfaction, and 5.9 on JSON schema adherence.
Trajectory as the Teacher: Few-Step Discrete Flow Matching via Energy-Navigated Distillation
Discrete flow matching can parallelize text generation across positions, but reaching high quality may require hundreds or thousands of model evaluations. The paper argues that few-step distillation fails primarily because teacher trajectories are built from blind stochastic jumps: an early poor midpoint decision propagates, and the student is trained to imitate it. Its Trajectory-Shaped Discrete Flow Matching (TS-DFM) uses a lightweight energy compass during training to score candidate continuations at each midpoint and select the most coherent path, without adding inference cost. In experiments with 170M-parameter language models, an eight-step shaped student achieves 32% lower perplexity than a 1,024-step teacher and runs 128 times faster, with gains across source distributions and three evaluators; it also outperforms the compared discrete-generation baselines, including systems using more data or larger models.
How energy teams turn theft detection into governed action with Genie and AI business processes
Energy theft detection often stops at an ML risk score, leaving revenue-protection teams to manually interpret cases, prepare investigations, dispatch crews, recover losses, and report results. The described Databricks workflow turns that score into a governed operational loop: Lakeflow prepares features, models run through Model Serving and Unity Gateway, and a Databricks App uses Lakebase to store live case state. AI-generated summaries explain why accounts were flagged, while dispatch-ready reports provide next steps, evidence checklists, and safety notes. Genie One answers leadership questions using metric definitions in Unity Catalog, and an Agent Bricks Multi-Agent Supervisor can assemble a traceable month-end board pack. Unity Catalog lineage, PII labels, access controls, model guardrails, and Unity Gateway routing support regulated use under frameworks including Ofgem’s and ELEXON’s requirements, while the same pattern is presented as applicable to other ML-driven interventions.
How Databricks’ marketers use data 3x more with Genie, an AI analytics assistant
Databricks marketers struggled to get trusted answers from fragmented marketing data quickly, prompting the creation of Marge, a marketing implementation of Genie Agents. The team first unified campaign, sales, web, event and advertising information in a governed Marketing Lakehouse, then grounded the assistant in business definitions, verified logic and example queries. Marge translates natural-language questions into analytical queries, uses Unity Catalog for lineage and role-based access, and can ask clarifying questions or perform multi-step analysis. The reported results include threefold greater data use in decisions, more than 85% adoption, 50% quarter-over-quarter growth, over 5,000 answered questions and a 25% reduction in flagged incorrect responses. Databricks attributes the rollout to continuous feedback, benchmark evaluation, focused use cases and embedding Genie in existing workflows rather than treating conversational analytics as a one-time deployment.
From zero-shot forecast to purchase order with Amazon Bedrock AgentCore
This post presents an inventory automation architecture that moves from zero-shot demand forecasts to validated purchase orders while applying safety stock, minimum order quantities, budgets, and promotional adjustments. It combines Amazon Chronos2, hosted on SageMaker Serverless Inference, with four Claude-powered agents built using the Strands Agents SDK and deployed on Amazon Bedrock AgentCore; deterministic Python tools perform computation and agents handle judgment. Product data and future covariates arrive through CSV files in S3, while business rules remain in JSON, enabling what-if forecasts and onboarding without per-SKU training. In internal testing across 50 SKUs over a four-week horizon, the system reached 12.3% median WAPE, reduced onboarding from 2–3 weeks to under five minutes, and lowered estimated monthly inference cost from about $1,091 to $15. A conditional retry loop surfaces budget violations for user decisions instead of silently truncating orders, with traces and evaluator scores supporting auditability.
DiscoSign: Discourse-Aware Text to Sign Language Gloss Translation
DiscoSign addresses a limitation in sign language processing: systems have traditionally translated at the sentence level, overlooking discourse phenomena important to sign language comprehension. It presents a modular Large Language Model (LLM)-based framework for text-to-sign language gloss translation that handles spatial coreference resolution, Question-Answer Clauses (QACs), and concept-gloss consistency, preserving entity locations and stable English-to-ASL sign mappings across discourse. Because traditional translation metrics do not measure discourse-level quality, the authors introduce metrics for each coherence dimension targeted by the framework. Experiments on sentence-level and discourse-level datasets report improved spatial consistency and entity tracking over sentence-only translation, while maintaining competitive single-sentence gloss quality; the work describes this as the first systematic framework with a corresponding evaluation methodology.
SimpleDesign: A Joint Model for Protein Sequence and Structure Codesign
SimpleDesign addresses protein design as a joint modeling problem involving amino acid sequences and three-dimensional structures, whose interplay determines protein function. Unlike approaches that first train autoencoders and then model their latent representations, it uses a single-stage, end-to-end objective directly in data space. The objective combines discrete cross-entropy for sequences with a regression objective for structures, while Transformer-based multimodal backbones provide modality-specific processing and global self-attention across both modalities. Trained on more than 2M sequence-structure pairs, SimpleDesign achieves competitive performance on co-design and unconditional sequence and structure generation benchmarks. The work argues, based on these results, that multi-stage training is not necessary for performant co-design models.
Halving the Time: How Uber Eats Rebuilt Its Search Pipeline
Uber Eats rebuilt its search pipeline after accumulated architectural decisions made latency increasingly costly across retrieval, hydration, ranking, ads, and presentation. The team changed its primary metric from backend API response time to Above-the-Fold completion, then introduced paginated responses with server-side caching and concurrent template rendering; these changes improved ATF by over 200 milliseconds. Other work removed low-yield retrieval strategies, grouped items by product identity to reduce lookups by over 100 times, split ranking and presentation hydration, eliminated false DAG dependencies, added request hedging, and redesigned ad data as column-oriented in memory. The combined effort cut search latency in half, including approximately 120 milliseconds from retrieval, 100-plus from hydration changes, and 130 from ads, while planned product-based search, zero-pass ranking, and streaming target further gains.
Refreshing the Travel-Time Map Behind Lyft’s Marketplace: Rebuilding Neighborhood Reachability…
Lyft’s Neighborhood Reachability Signals dataset supplies precomputed travel-time estimates between geohash-6 cells and a vocabulary of usable cells for pricing, driver heatmaps, and marketplace features. Production ETAs had remained largely based on 2018–2019 data, systematically underestimating travel times and omitting genuinely matchable pairs, while consumers maintained separate allowlists and denylists. The FORTOP team redesigned the Airflow-generated files by rebuilding the geohash universe from demand and supply history, estimating dense pairs from trip history, using Routing Simulation System (RSS) fallbacks for sparse pairs, and blending partial observations. In the second half of 2025, the refreshed dataset was adopted by Pricing after validation, with reported improvements in accuracy and completeness, a six-month refresh cadence, and plans for nine time-category files per region to represent traffic patterns.
Top topics
- Architecture832
- Performance814
- LLMs658
- Machine Learning620
- Open Source563
- AI Agents549
- AI465
- Developer Experience448
Weekly digest
The best engineering reads, once a week.