Open-source reference implementations of the systems modern products actually run on — AI-native marketplaces, agent copilots, enterprise integrations, and data platforms. Deployable, real, small enough to read end-to-end.
Every repository below is a working reference implementation of a pattern that actually ships — small enough to read end-to-end, complete enough to deploy with real credentials. Open the case studies to see the architecture, the engineering decisions, and the trade-offs.
AI-native managed marketplace for home services — one app for plumbing, electrical, cleaning, painting, beauty, appliance repair. Photo + sentence → diagnosed JobScope → transparent quote → vetted pro, with leakage defense at the core.
A self-improving educational-content engine. An agent grounds lesson generation in validated context packs, scores it with a hybrid eval harness, then rewrites the system itself — its own prompts, its reward weights, and new check-tools it designs — keeping only changes that beat a frozen north-star without regressing a held-out set.
Payment and customer events into reliable multi-system automations. Signature verification, idempotent ingestion, queue-backed execution, per-step retries, encrypted credentials, and an operator console.
Plugin-driven data movement across AWS, Azure, GCP, Snowflake, Postgres and Oracle. Each connector implements the same five-method interface; jobs configured via UI wizard or YAML.
Patient data from FHIR R4, legacy Oracle EHR, lab SFTP drops, and EDI 837/835 reconciled into a master patient index with probabilistic matching, Safe-Harbor masking, and HIPAA audit.
Self-hosted EMEA accounts-payable automation that replaces $100K/yr SaaS. Ingest → extract → rules → AI confidence → human review → 27-country VAT → BI, in one auditable transaction.
Vendor and contract documents into structured, evidenced, risk-scored decisions. Replaces $100K/yr SaaS (Vendr, Ironclad) with a self-hosted policy engine, human-in-the-loop queue, and immutable audit log.
Autonomous agent that reads every onboarding signal — Slack, Jira, email, calls, docs, CRM — and produces evidence-cited briefs, calibrated slip-risk forecasts, and review-ready customer emails. Catches slips before they happen, with every claim grounded in a verifiable source.
AML alert-triage and SAR-drafting agent that scores every alert on a continuous suspicion scale, cites each typology claim to the exact transaction behind it, and drafts a regulator-ready SAR — clearing false positives while never letting a true positive slip. The agent proposes; a compliance officer disposes and files.
AI front-desk for the trades that turns an inbound call or text into a booked, dispatched job — triaging the problem, quoting a transparent price band from the business's own price book, and assigning the nearest qualified tech by an explainable scorer, with a non-skippable safety-escalation branch that never mistakes a gasket for a gas leak.
Post-purchase resolution agent that reads an inbound "where's my order / this arrived cracked / I want a refund", pulls the real order and carrier facts, cites the governing policy clause, and resolves the case end-to-end — tracking answer, exchange, carrier claim, save-the-sale offer, or refund — with a scored return-abuse guardrail and a human gate on anything that moves money.
Legal client-intake and matter-triage agent that qualifies a lead, runs a recall-first conflict check, tracks the statute-of-limitations clock from a content-hashed rule pack, and books or routes the matter — with every outbound gated on a TCPA consent ledger so the firm never contacts anyone without provable permission.
Healthcare prior-authorization agent that de-identifies the chart, decomposes the payer's own policy into criteria, judges each one met / unmet / unknown against cited clinical evidence, and assembles a submission-ready packet — predicting approval probability on a calibrated scale, drafting the appeal when a denial is overturnable, and gating everything behind a clinician attestation.
Daily financial ETL that pulls core banking, customer master, partner files, and FX rates into a Snowflake warehouse with fraud scoring, SCD-2 dimensions and SOX controls applied along the way.
Lambda architecture over Kafka — speed layer for sub-second sessionisation and bot scoring, batch layer for nightly CLV and funnel rollups, both over the same source-of-truth event log.
Kimball-style Snowflake warehouse unifying SAP ERP, Oracle WMS, MQTT truck IoT, and carrier APIs into conformed facts, SCD-2 dimensions, and an OTIF / fill-rate KPI engine.
The kinds of problems these systems are built to solve — across AI products, integrations, automation, and data.
Zero-to-one builds with AI at the core — intake, agents, human-in-the-loop, leak-resistant marketplaces.
LLM copilots grounded in real data — multimodal, citation-backed, with guardrails.
Salesforce, NetSuite, Stripe, Jira, EDI, FHIR, HL7 — connected reliably with replay and audit.
Manual multi-step processes replaced with durable workflows — triggers, retries, HITL gates.
Warehouse, lake, or lakehouse — greenfield builds, streaming, or legacy ETL migrations.
System design and the path from prototype to production — the bar for what "production" means in a codebase.
Every repository is MIT-licensed — fork it, lift the patterns, and ship it in your stack. The case studies above walk through the architecture and the decisions; the code is all on GitHub.