In this blog post I am talking about our Partner, Supercog, a West Coast US startup, focused on creating AI Assistants to simplify and improve existing digital environments. As illustrated below, this becomes what industry analyst Vernon Keenan cites as an example: Overlay AI¹ to Salesforce.

As an Overlay AI to Salesforce, Supercog is distinctively different to Agentforce, which is an Embedded AI. The advantages with Supercog are: prebuilt Tools with a wide variety of apps that work with Salesforce; choice of Large Language Model (LLM); and, flexible business value pricing.

Your CRM app could be HubSpot instead of Salesforce. We can AI-enable your everyday processes and tasks that combine CRM with say, a messaging app, project management, documents and email, as described below.

Today, Supercog works with a growing list of popular, world-class technologies, beyond Salesforce and HubSpot CRM, including: Amazon Web Services (AWS); Atlassian Jira; Google Workspace; Intuit Quickbooks; LinkedIn; Postgres; ServiceNow; Snowflake, and many more.

This is AI that works.

Everyday Solutions

Simply put, Supercog enables knowledge-intenisive organisations to increase productivity. It accomplishes this by creating AI Assistants that learn from any source of information, whether struictured or unstructured data.

So, Supercog frees up time for your team to spend more time focusing on what matters. You move faster with an AI Assistant powered by Supercog at work. It's really as simple as saying 'right information, right app, right person, right time'.

Use Cases for AI Assistants can be designed as measurable value outcomes applied to everyday processes. Examples that could represent priorities for improvements in your organisation include:

• Analysing information related to high-value, high-touch sell environments, where an AI Assistant can extract key insights from say, a Salesforce or HubSpot CRM system, Slack messaging, Google Doc or Word and PDF documents, Google Slides or PowerPoint presentations plus emails.
• Accelerating hiring and onboarding in knowledge-intensive industries, such as financial services or healthcare, where an AI Assistant can verify candidate credentials. licences and certifications to ensure they meet regulatory requirements.
• Generating billings forecasts from bookings forecasts, where an AI Assistant extracts data from structured sources, such as spreadsheets and unstructured sources, such as proposals and email threads that enrich CRM data.
• Automatically creating and assigning project tasks, where an AI Assistant can continuously optimise resource allocation across projects and flag delays or bottlenecks to enable faster, proactive human-in-the-loop interventions.
• Transcribing meetings held in Google Meet, Microsoft Teams or Slack Huddle, where an AI Assistant can generate a concise summary plus identify actions and timelines, where tasks could then be automatically created and assigned in Atlassian Jira.

Supercog AI Assistants can work with any LLM, as described below. Today, GPT4o-mini is proving most cost-effective with our engagements, where we amortise the cost of Tokens consumed as predictable, fixed-price outcomes. However, DeepSeek is now proving effective for us too.

Success Stories

At Being Guided, I work with Supercog to identify and engage early adopter customers within the vast and diverse Salesforce customer base. In recent months together, we have delivered the Supercog AI to the UK National Health Service (NHS), Europe's largest employer and Arts Care, a growing pioneer of social prescribing services in Northern Ireland.

For the NHS, Supercog has performed an AI Augmentation role as Copilot in designing and developing the native integrations between Salesforce (Apex Code, Lightning Web Components) and Electronic Health Record (EHR) and Patient Appointment Booking systems, including Health Level Seven (HL7) data exchange standards.

The work with Arts Care and Supercog has resulted in the creation of automated Scheduling and Invoicing for a growing ecosystem of artists delivering social prescribing sessions to patients in both hospital and community settings throughout Northern Ireland. As with NHS, Supercog has acted in an AI Augmentation role as Copilot for creating Rules and Triggers related to Invoicing Cycles.

How It Works

Supercog AI Assistants are built on AI Agents, which are semi-autonomous programs that follow simple English instructions. Each Agent is powered by a Large Language Model (LLM), which gives it the ability to understand language, make plans and execute tasks.

The Supercog AI Agent Platform consists of three elements: Large Language Model (LLM); Agent Instructions; and, Tools. Here's brief explanation of how they come together to deliver AI Assistants that extend your Salesforce environment:

Large Language Model (LLM)
The LLM is powering the Agent. You can choose amongst many different LLMs, including Google Gemini, GPT4o-mini; Anthropic Claude; and, since mid-January 2025, DeepSeek.

Agent Instructions
The Agent Instructions describe the work that you would like the Agent to perform. If you want to devote your Agent to a specific task, then you will want to explain that task in the Instructions.

Tools
We enable Agents to perform work by giving them Tools. Templated Tools currently available are: Amazon S3; Atlassian Jira; Discord; DuckDB; GitHub; Google Calendar; Google Docs; Google Gmail; HubSpot; Hunter.io; Intuit Quickbooks; REST API; SMS; Salesforce; ServiceNow; Slack; Snowflake; YouTube; and, Zapier.

Licences and Services Fees

Supercog Licences
£49 per per User per Month. Subscription Fees paid annually in advance. Includes Support Services. Volume discounts apply.

Managed Services
Fixed-price offers. Let's setup a Discovery Workshop to scope-out your first AI Assistant. Professional Services Fees per Engagement, paid on pre-agreed Use Case and Deliverables.

Reference

1. Keenan, V. (2024) SalesforceDevOps.
   https://salesforcedevops.net/