COMPUTER SCIENCE

The fundamentals
under the work.

Good software rests on computer science: algorithms, data structures, systems, and security. Here is how those fundamentals show up in client work.

6Core disciplines we apply
O(n log n)Typical target for heavy paths
ACIDDefault for money and records
0Secrets in source code

TOPIC INDEX

Disciplines
in practice.

Open a topic to see where it matters in real systems.

Algorithms and complexity

Choosing the right approach for scheduling, routing, search, and matching so systems stay fast as data grows.

SchedulingRoutingSearch
Data structures

Indexes, queues, caches, and trees selected for how data is read and written in practice.

IndexesQueuesCaches
Distributed systems

Consistency, retries, idempotency, and failure handling across services and providers.

IdempotencyRetriesConsistency
Databases

Relational modeling, transactions, row-level security, and query plans that hold up under load.

PostgreSQLTransactionsRLS
Security and cryptography

Authentication, encryption, key management, and threat modeling applied to each system.

OAuthTLSThreat models
Machine learning

Evaluation, retrieval, classification, and the limits of language models in production.

EvaluationRetrievalClassification

APPLIED

Theory, applied.

EventDedupe keyHandlerRecordfeedback loop

DISTRIBUTED SYSTEMS

Exactly-once results from at-least-once delivery.

Payments and messages are retried by every provider. We design idempotent handlers so a retry never charges or texts someone twice.

Request with company IDPolicy checkRows for that company only

DATABASES

Tenancy enforced by the database.

Row-level security keeps each company's data separate even if application code has a bug.

Want a second opinion on an architecture decision?