Skip to main content
Subscribe
EDITORIAL DESK ARCHIVE

AI Workflows

Step-by-step production AI workflow architectures, event loops, and agent orchestration for builders.

Deep Dive AI Workflows

Build an Always-On Personal Agent Workflow with Cloud Continuity & Spend Caps

Gemini Spark dropped from the $99.99 Ultra tier to the $19.99 AI Pro plan on July 25, 2026, making always-on personal agents a consumer product. This workflow builds spark-sleeper, a LangGraph pipeline that runs a personal agent 24/7 on cloud compute: it ingests tasks from any device, wakes on schedule or trigger, executes long-running work with a durable task queue, checks every spend against a hard monthly cap, and reports results back to the user's channel of choice — the Gemini Spark pattern as an open workflow you control.

Deepak Bagada Deepak Bagada
11m read
Deep Dive AI Workflows

Build an Agentic Patient-Journey Voice Workflow with Human Escalation Gates

Assort Health raised a $120M Series C at a $1.2B valuation in August 2026, expanding from voice AI into an agentic system for the whole patient journey. This workflow builds care-flow, a LangGraph pipeline for healthcare front offices: a voice agent answers inbound calls, triages the request with structured symptom intake, books or reschedules appointments against the practice schedule, and routes anything clinically uncertain through a human escalation gate before any medical decision.

Deepak Bagada Deepak Bagada
11m read
Deep Dive AI Workflows

Build an Outbound Voice-Agent Workflow with IVR Navigation & Call Completion Tracking

Google's agentic shopping features now place real calls to stores, and consumer callers like Assindo navigate IVR menus and hold queues. This workflow builds call-runner, a LangGraph pipeline that makes outbound calls for a user: it plans the call from a task brief, dials through Twilio, navigates IVR menus with speech recognition, handles hold queues, executes the script, and logs a completion verdict — done, needs callback, or failed — with a transcript the user can check.

Deepak Bagada Deepak Bagada
11m read
Deep Dive AI Workflows

Build a Least-Privilege Agent Sandbox Workflow with OS Isolation

Hazmat (open-source, ~Aug 17 2026) runs AI coding agents in a dedicated least-privilege OS account so they cannot read SSH keys or cloud credentials, backed by a TLA+ formal spec, per-session macOS backups, and a session-scoped firewall. This dispatch builds agentcell, a LangGraph workflow that plans a session with declared paths, generates a per-session sandbox policy, preflights a backup, spawns any harness (Claude Code, Codex, OpenCode, Cursor) inside the account, and terminates on the first escape attempt.

Deepak Bagada Deepak Bagada
11m read
Deep Dive AI Workflows

Build a Capability-Tier Workflow with Cross-Model Subagents

OpenAI shipped cross-model delegation in Codex Multi Agents v2 (Aug 15, 2026): GPT-5.6 Sol orchestrates while delegating narrow tasks to cheaper GPT-5.6 Luna pure subagents that cannot spawn more agents, with guidance (Provencher) capping fan-out at 6-8 subagents. This dispatch builds modelrouter, a LangGraph workflow implementing capability-tier routing: an orchestrator decomposes a job, routes each task to the cheapest capable tier, enforces a max subagent count, and uses a quality gate to re-escalate failed tasks to a higher tier.

Deepak Bagada Deepak Bagada
10m read
Deep Dive AI Workflows

Build a Machine-Payment Workflow with Human Spend Ceilings

Cloudflare Agents Week (Aug 4-15, 2026) launched Cloudflare Wallets and cloudflare.pay on the x402 protocol so agents can hold verifiable identity and purchase within human-set limits, plus an Identity-Aware AI Gateway (Aug 5, 2026) attaching verified identity to every outbound AI request. This dispatch builds machpay, a LangGraph payment workflow where an agent requests a purchase, the workflow validates the request against a human-set spend ceiling, checks wallet balance, settles via x402 in a single idempotent request, and logs every spend line to an audit ledger with identity attribution.

Deepak Bagada Deepak Bagada
10m read
Deep Dive AI Workflows

Build a Regulated Account-Opening Agent Workflow with Human Approval Gates

Zeplyn's Schwab integration (Aug 12, 2026) showed agentic account opening cuts NIGO by 80% — but regulated account opening needs compliance rails, not just automation. This workflow builds account-open, a LangGraph pipeline that drafts the entire account-opening workflow, runs KYC/AML and eligibility checks, routes every uncertain field to a human approval gate, and writes an immutable audit trail before submission. It is the assisted-automation pattern for regulated finance.

Deepak Bagada Deepak Bagada
11m read
Deep Dive AI Workflows

Build an Explainable eDiscovery Fact-Investigation Workflow with Citation Trails

DISCO's Advanced Research GA (Aug 13, 2026) proved agentic eDiscovery can investigate facts autonomously — but legal defensibility demands citations on every finding. This workflow builds fact-finder, a LangGraph pipeline where an investigator agent plans a line of inquiry, runs multi-step searches across a document corpus, assembles findings with per-claim citation trails, and routes the trail through a human review gate before anything is admitted as a conclusion.

Deepak Bagada Deepak Bagada
11m read
Deep Dive AI Workflows

Build a Cross-Chain Agent Settlement Workflow with Approval Gates & Audit

Chainlink for Agents (Aug 14, 2026) made CCIP cross-chain settlement a first-class agent capability — which means every agent transfer is now a governance decision. This workflow builds settle-bot, a LangGraph pipeline that sits between an agent and its money: it verifies the transfer against verified price feeds, checks per-transaction caps and policy, routes value movement through a human approval gate, and writes an immutable audit record before and after settlement.

Deepak Bagada Deepak Bagada
11m read
Deep Dive AI Workflows

Build a Civic Service-Delivery Agent Workflow for Public-Impact Deployments

Code for India's Code for a Billion — Bharat Agentic-AI Hackathon 2026 (Aug 15, 2026) targets education, health, climate, governance, and financial inclusion. This dispatch builds civic-serve, a LangGraph workflow that turns public-service intents into governed agent actions: intake plus eligibility check, privacy-gated data access, multilingual output, human approval for entitlement-changing actions, and an append-only audit for public-sector transparency compliance.

Deepak Bagada Deepak Bagada
11m read
Audio Briefing
Accessibility Preferences
High Contrast Mode
Accessible Reading Font

Keyboard Shortcuts

Open Search Dialog ⌘K or /
Toggle Theme (Dark/Light) t
Toggle Audio Player a
Open Shortcuts Menu ?
Close Active Dialog Esc

Cookie & Privacy Preferences

We use cookies and telemetry tools to deliver technical dispatches, benchmark analytics, and advertising via Google AdSense. Review our Privacy Policy.