Skip to content

Founder-led. Limited engagements.Talk to Us →

Journal

Notes from building systems

What works, what breaks, and what we learned building automation systems, Agentforce agents and products. Working notes, not content written for traffic.

Latest · 20 Nov 2026

Agentforce vs Copilot Studio vs Custom AI Agents (and Zoho Zia)

Agentforce, Copilot Studio, a custom agent or Zoho Zia: the right choice follows your system of record. A side-by-side comparison with list prices, a worked cost example and India pricing in INR.

20 Nov 2026 · 6 min

AI Document Chat System | 30 Days of Building AI Systems, Day 2

19 Nov 2026 · 8 min

Agentforce POC: What a Two-Week Pilot Should Prove

19 Nov 2026 · 8 min

[Rewrite] Agentforce Readiness Assessment: 12 Checks Before You Build

Series

30 Days of AI Agents

A 30-day series taking AI agents from first principles to production: architecture, memory, tools, frameworks, RAG, evaluation, safety and scaling.

30 Days of Building AI Systems

A 30-day build series: one real-world AI system per day, from research assistants to multi-agent platforms, with architecture, stack and production notes.

Earlier notes
  1. 19 Nov 2026

    Salesforce Agentforce: What It Does, Costs, and When It Fits

    A plain guide to Salesforce Agentforce in 2026: how an agent routes, plans and acts, what Flex Credits, conversations and per-user licences cost, and where a Flow or a custom agent is the better answer.

  2. 19 Nov 2026

    AI Research Assistant System | 30 Days of Building AI Systems, Day 1

    Build an AI research assistant: web search, RAG retrieval, and LLM synthesis into a sourced report, with architecture, stack, evaluation, and failure modes.

  3. 18 Nov 2026

    The Future of AI Agents: What's Next? | 30 Days of AI Agents, Day 30

    The 30 Days of AI Agents finale: a recap of the whole series, the trends shaping agents next, what builders should focus on, and the skills that will matter.

  4. 17 Nov 2026

    Scaling AI Agents to Production | 30 Days of AI Agents, Day 29

    Scaling AI agents to production: architecture, readiness checklist, state, failure handling, security, cost controls, and the signals to monitor under load.

  5. 16 Nov 2026

    Evaluating Multi-Agent Systems | 30 Days of AI Agents, Day 28

    How to evaluate multi-agent systems: end-to-end task success, handoff quality, coordination failures, loops, and cost and latency attributed to each agent.

  6. 15 Nov 2026

    Fine-Tuning AI Agents | 30 Days of AI Agents, Day 27

    When and how to fine-tune AI agents: training data that works, LoRA and QLoRA, tool-call examples, evaluation before and after, and when prompting and RAG win.

  7. 14 Nov 2026

    Multi-Agent Deployment Strategies | 30 Days of AI Agents, Day 26

    How to deploy multi-agent systems: centralized, distributed, microservices, serverless and hybrid strategies, plus queues, idempotency, retries and a checklist.

  8. 13 Nov 2026

    AI Agent Observability | 30 Days of AI Agents, Day 25

    AI agent observability in practice: what to log, trace and measure, an OpenTelemetry span example, LLM cost and latency metrics, alerts, and common pitfalls.

  9. 12 Nov 2026

    AI Agent Memory Optimization | 30 Days of AI Agents, Day 24

    Optimize AI agent memory for cost and quality: context-window budgets, summarization, compression, relevance and recency scoring, TTLs, and retrieval checks.

  10. 11 Nov 2026

    Advanced RAG Techniques | 30 Days of AI Agents, Day 23

    Nine advanced RAG techniques explained: hybrid search, query rewriting, multi-query, reranking, compression, parent docs, step-back, time-aware, agentic RAG.

  11. 10 Nov 2026

    AI Agent Safety & Guardrails | 30 Days of AI Agents, Day 22

    A practical guide to AI agent guardrails: input, action, output, context and workflow layers, prompt injection defense, least privilege, and human escalation.

  12. 9 Nov 2026

    Evaluating AI Agents | 30 Days of AI Agents, Day 21

    How to evaluate AI agents in practice: test sets, metrics by category, LLM-as-a-judge, human review, benchmarks, and a continuous evaluation loop that sticks.

  13. 8 Nov 2026

    Multi-Agent Collaboration Patterns | 30 Days of AI Agents, Day 20

    Six multi-agent collaboration patterns explained: supervisor, peer-to-peer, hierarchical, pipeline, blackboard, swarm, and how to choose the right one.

  14. 7 Nov 2026

    AI Agent Planning & Reasoning | 30 Days of AI Agents, Day 19

    How AI agents plan and reason: task decomposition, reasoning types, ReAct and Reflexion, a travel planner example, and a plan-execute-replan loop.

  15. 6 Nov 2026

    AI Agent Tools | 30 Days of AI Agents, Day 18

    How AI agent tools and function calling work: the tool loop, tool types, writing tool schemas, a safe executor in code, best practices and failure modes.

  16. 5 Nov 2026

    Multi-Agent Systems | 30 Days of AI Agents, Day 17

    Multi-agent systems in depth: orchestrators, system types, communication patterns, shared memory design, a CrewAI example, and how to keep them reliable.

  17. 4 Nov 2026

    AI Agent Memory | 30 Days of AI Agents, Day 16

    AI agent memory in depth: memory architectures, write and retrieval policies, summarization, forgetting, storage options, and the metrics to track.

  18. 3 Nov 2026

    RAG (Retrieval-Augmented Generation) | 30 Days of AI Agents, Day 15

    RAG explained step by step: ingest, index, retrieve, augment, generate. Architecture, a minimal implementation, RAG vs fine-tuning, and best practices.

  19. 2 Nov 2026

    LangGraph | 30 Days of AI Agents, Day 14

    LangGraph explained: build stateful agent workflows with nodes, edges, conditional routing and shared state, with a working Python example and best practices.

  20. 1 Nov 2026

    Vector Databases | 30 Days of AI Agents, Day 13

    Vector databases explained for AI builders: embeddings, similarity search, HNSW and IVF indexes, choosing Pinecone, Qdrant, pgvector and more, and RAG.

Subscribe to the edition

One useful systems note a week from real Salesforce and automation builds. Unsubscribe in one click.