From Clicks to Conversions: Architecting Shopping Conversion Candidate Generation at Pinterest (7 minute read)
Pinterest built a dedicated two-tower retrieval model to generate better shopping ad candidates optimized for offsite conversions, moving beyond traditional click/engagement-based signals which are abundant but poorly correlated with actual buying intent. The system uses a unified multi-task architecture with parallel DCN v2 and MLP cross layers, clever training techniques to handle sparse and noisy conversion data, and an advertiser-level loss function.
|
How Vinted Serves Personalised Search Autocomplete (9 minute read)
Vinted rebuilt its search autocomplete system, moving from static, generic suggestions to a hybrid approach combining a strong heuristic scoring model with a Learning-to-Rank (LTR) model. They score suggestions offline using popularity, sell-through rate, and usage signals, index them with clever prefix and fuzzy matching techniques, then apply a LightGBM model in real-time that incorporates user behavior and context to re-rank results.
|
Skipper: Building Airbnb's embedded workflow engine (12 minute read)
Skipper is a lightweight, embedded workflow engine designed to provide durable and reliable execution for long-running business processes (like insurance claims and payments). Instead of relying on external orchestration tools or queues, Skipper uses a simple annotation-based approach to persist state in the service's existing database and achieves durability through deterministic replay.
|
|
GraphRAG beyond the demo: Lessons from the trenches (12 minute read)
GraphRAG is most useful when questions require multi-hop reasoning across documents, entity relationships, or system-level dependencies: use Vector RAG for simple factual lookups and keep GraphRAG as an opt-in backend. In production, the main pain points are heavy indexing cost, difficult updates, multi-layer evaluation, and infrastructure that usually needs batch jobs rather than request-path execution. Success depends on selective graph scope, explicit update policies, repeatable evals, and strong observability/cost controls.
|
A/B Testing Pitfalls: What Works and What Doesn't with Real Data (5 minute read)
A/B testing failures are far more often caused by broken infrastructure and poor experimentation practices than by the ideas being tested. Common failures include Sample Ratio Mismatch (SRM) from bad randomization, early peeking that inflates false positives, insufficient statistical power, and optimizing the wrong metrics without guardrails, causing misleading results.
|
|
Rocky (GitHub Repo)
Rocky is a Rust-based tool that adds a control layer on top of data warehouses, helping teams manage pipelines with features like data contracts, lineage tracking, and safe testing through branches. It focuses on catching errors early, preventing data issues, and making data workflows more reliable and easier to understand.
|
oLLM (GitHub Repo)
oLLM is a Python library for running very large context LLM workloads on modest consumer hardware by offloading model weights and KV cache to SSD instead of keeping everything in GPU memory. It's useful for offline tasks like analyzing long documents, logs, contracts, chats, or reports locally without quantization.
|
HOT Updates in Postgres (12 minute read)
HOT Updates in PostgreSQL is a clever storage optimization that allows UPDATEs on unindexed columns to avoid touching indexes entirely when the new tuple fits on the same page as the old one. Instead of creating new index entries, PostgreSQL marks the old tuple as HOT_UPDATED and places a HEAP_ONLY tuple on the same page, forming a chain that scans can follow, which reduces WAL traffic, index maintenance, and vacuuming overhead.
|
Materialized Tables in Apache Flink (14 minute read)
Materialized Tables in Apache Flink allows users to define a table directly with its population query, embedding both the schema and the continuous or scheduled refresh logic inside the catalog. This simplifies ETL pipelines by automatically handling job lifecycle, schema evolution, and refreshes.
|
|
Running SQLite in the browser with sql.js and WASM — a practical guide with Google Drive sync (5 minute read)
A client-side architecture uses SQLite compiled to WebAssembly in the browser, with the database persisted as a single binary file on the user's Google Drive. Compared with IndexedDB or proprietary sync layers, this gives true data portability and privacy: the file can be opened in any SQLite tool, while Drive access is limited via the drive.file scope. Local state is written to localStorage after each mutation, Drive sync is debounced by 10 seconds, and conflict handling prefers Drive as the source of truth.
|
Building a High-Scale Real-Time Recommendation Engine with Feature Stores and Redis Observability (5 minute read)
Real-time recommendation systems now need to combine rich contextual features with sub-100 ms latency at scale, often across billions of interaction records. Feature stores act as the consistency layer between offline training and online serving, reducing training-serving skew, while batch platforms compute expensive features and embeddings. Redis is used for low-latency vector similarity search, candidate retrieval, and caching eligibility filters, keeping request paths fast and efficient.
|
|
How Linux 7.0 Broke PostgreSQL (9 minute read)
Linux 7.0 accidentally cut PostgreSQL performance in half because a scheduling change increased how long spinlocks were held during memory page faults, causing massive CPU waste, and switching to huge memory pages fixes the issue.
|
|
Love TLDR? Tell your friends and get rewards! |
|
Share your referral link below with friends to get free TLDR swag!
|
|
|
|
| Track your referrals here. |
|
|
|
0 Comments