Salesforce Data Cloud Consultant and AI Agent Engineer — Practical AI Consulting for Small Businesses
I work hands-on with small and mid-sized companies to figure out what's slowing your team down — and whether technology can actually help. Sometimes that means building an AI agent. Sometimes it means wiring up tools you already have. Often it means amplifying the ideas of people already inside your company whose input isn't making it to the table. When I leave, your team should be more capable than before — not more dependent.
Schedule a free consultation · View projects
How I work
Before I recommend any tool, I spend time with your team to understand how work actually gets done. I focus on AI for businesses without a tech team — no in-house AI expert required. Sometimes the answer is AI. Sometimes it isn't. Either way you'll know — and you'll have someone who can build it with you if it is.
- I listen first. The right answer for your business might not be AI — and if it isn't, I'll tell you that.
- I work with what you have — Salesforce consulting, custom tooling, or whatever already fits your team.
- I amplify the people already inside your company. Often they already know what needs to change.
- When I leave, your team is more capable than before — not more dependent on me.
Three ways I work with you
Not every problem needs AI. The hard part is knowing which ones do. Sometimes it's one of these. Sometimes it's all three in sequence. We figure out which makes sense together — usually on the first call.
Figure out where AI fits in your business
A structured review before you commit budget.
Not every problem needs AI. The hard part is knowing which ones do. We spend a couple of weeks looking at how your team actually works, and you end up with a prioritized plan you can act on — with or without me.
- Workflow audit — what's broken, what's slow, what's tolerated
- Honest read on where AI helps and where it doesn't
- Prioritized roadmap with time and cost estimates
- Review of what your team can build themselves vs. needs help with
Build AI tools that fit how your team already works
Custom agents for the parts of your business that actually need them.
For the problems where AI genuinely helps, I build custom agents that connect to your existing systems, remember what matters, and hand off to a human when they should. Deployed on Salesforce, in a Slack workspace, or on your own server — wherever your team already is.
- Automate repetitive work without replacing your people
- Connect systems that don't currently talk to each other
- Build assistants your team can actually trust
- Keep everything auditable — you always know what the agent is doing
Connect your data so your team stops doing it by hand
For teams on Salesforce — or wondering if they should be.
If your business already runs on Salesforce, my Salesforce consulting work focuses on Data Cloud and Agentforce so sales, service, and marketing work from a single view of the customer. If you're not on Salesforce, we figure out together whether you should be, or whether something simpler fits.
- Unify customer data across the systems you already use
- Set up identity resolution so one customer is one record
- Build segments your marketing team can actually act on
- Activate data to Marketing, Commerce, or Service Cloud
Things I've built
A mix of tools I've shipped for clients and things I've built for myself. Each one started with a real problem — either mine or someone else's — not with a search for somewhere to apply AI.
AI memory system for agents
An AI memory system — an assistant that remembers everything you've told it, across months of conversations — so your team stops re-explaining the same context every time. Built it for my own consulting work, now open source and used in production daily. 540+ tests covering the core. Works with Claude Code, Gemini, and any MCP client.
Salesforce Data Cloud knowledge engine
An LLM-maintained reference for teams implementing Salesforce Data Cloud — automatically scored for accuracy, so you're not relying on outdated docs or generic AI answers when you hit a real edge case. 895/895 across 9 benchmarks, 500+ indexed chunks from real implementations.
Inventory health dashboard
A real-time view of what's stocked, what's running low, and what's anomalous — with automated alerts before items run out. Deployed on Salesforce Managed Runtime for a retail team that used to reconcile three separate spreadsheets by hand.
Salesforce security benchmark
An automated tool that checks 25 security controls against any Salesforce org — so teams don't need to do manual audits every quarter. Open source, shipped as a Pi agent package.
Slack message triage
Incoming messages in a busy Slack workspace get classified and routed to the right person automatically — cutting the "who handles this?" friction that slows teams down when volume picks up. Built on n8n, no custom infrastructure.
AI news aggregator
Collects AI news from four source types, removes duplicates, and surfaces what's actually new. Built because I was losing an hour a day to ten overlapping newsletters saying the same thing.
Education & Certifications
Education: Grado en Ciencia de Datos e Inteligencia Artificial — VIU, Universidad Internacional de Valencia (in progress, Sept 2024–present). Project Management (PMI aligned), Universidad Piloto de Colombia, 2014.
Salesforce consulting: Certified Data Cloud Consultant · Certified Business Analyst · Certified AI Associate · Certified Associate.
Management: PMP – Project Management Professional (PMI) · Agile (Scrum/Kanban) expertise.
Technical & Engineering: KeepCoding Java Backend Development · Boot.dev Python coursework (Learn to Code, OOP, Asteroids, Bookbot) · Mobile Test Automation (Appium, Cucumber, Ruby).
Frequently Asked Questions
What is a Salesforce Data Cloud consultant?
A Salesforce Data Cloud consultant designs and implements unified customer data platforms within the Salesforce ecosystem. They configure data stream ingestion, build identity resolution rules, create calculated insights, and activate segments across Marketing Cloud, Commerce Cloud, and Service Cloud — turning siloed data into a single, actionable customer profile.
What is an MCP server and why does it matter for AI agents?
MCP (Model Context Protocol) is an open standard that lets AI models securely connect to external tools and data sources. An MCP server exposes structured capabilities — like database queries, API calls, or file operations — that an AI agent can invoke at runtime. This matters because it gives agents reliable, auditable tool use instead of brittle prompt-based workarounds.
How does AI agent memory work?
An AI memory system stores and retrieves context across sessions using a combination of full-text search (FTS5), vector embeddings, and structured metadata. When an agent starts a new session, it queries its memory layer for relevant past interactions, decisions, and facts — enabling continuity without re-explaining context every time.
What is Agentforce and how can a consultant help?
Agentforce is Salesforce's platform for building and deploying autonomous AI agents within the Salesforce ecosystem. A consultant helps by assessing your org's readiness, designing agent topics and actions, connecting Data Cloud for grounding, and establishing guardrails — ensuring agents operate safely within your business rules and compliance requirements.
Can AI agents integrate with existing Salesforce orgs?
Yes. AI agents connect to Salesforce through standard APIs, MCP servers, and platform events. They can read and write CRM data, trigger flows, classify incoming cases, and automate multi-step processes — all while respecting your existing permission model and security controls.
What industries benefit from AI agent engineering?
Any industry handling complex data workflows or high-volume customer interactions benefits from AI agent engineering. Retail and CPG companies use agents for inventory intelligence and demand forecasting. Financial services automate compliance monitoring. Healthcare organizations streamline patient data routing. The common thread is structured data, repetitive decisions, and a need for speed.