Tell me what you are trying to build, fix, govern, prove, or launch, and I will point you to the public Folium page that fits. It uses public routes only, so do not send private data here.
Public-facing PDF Review before production Folium Systems
Investor pitch deck
Folium Systems Investor Pitch Deck
This deck gives investors and strategic reviewers a boardroom-readable view of Folium Systems without inventing financial metrics or exposing private diligence. It is record-led, public-facing, and built for a controlled first conversation.
- Audience
- Qualified investor conversations, strategic partners, executive reviewers
- Purpose
- Frame Folium's AI operating-capability thesis and the diligence questions that should follow
- Updated
- June 2026
- Use it to decide Whether this is an education, audit, first-build, pilot, trust-review, or operations conversation.
- Keep gated Private data, credentials, customer records, live providers, regulated authority, and production dependency stay outside public review.
- Bring to the room One workflow, one owner, the systems it touches, the records involved, and the decision leadership needs to make.
- Folium is positioned in the scarce layer between AI access and operating capability.
- The deck answers the expected investor spine: why now, problem, customer, solution, product, traction signals, model, moat, team, capital use, and risk.
- The digital manufacturing plant is the platform: reusable records, software, trusted-data utilities, agents, governance, and launch systems.
- The next diligence step is to inspect the engine and keep financial, customer, and offering claims in controlled materials.
Pitch spine
The deck is strongest when the engine is inspectable.
A serious pitch should show why now, the customer pain, Folium's wedge, the productized delivery plant, the trust posture, and the diligence path without inventing numbers.
Boardroom read
- Why now
- Problem
- Solution
- Moat
- Diligence
01Gives the deck a boardroom spine without unsupported financial theater.
02Connects market pain, productized delivery, moat, and diligence into one read.
03Keeps the story ambitious while staying honest about public-facing boundaries.
R
Navigation map
Choose the review route before reading cover to cover.
This packet is meant to support a real decision meeting. Let each reviewer enter through the route that matches their job, then bring the group back to the same controlled next step.
- Decision route
- Operating route
- Trust route
Executive route
Decision first
Start with the cover, visual summary, executive read, controls, first ninety days, and handoff. This route helps leaders decide whether the next move is education, audit, first build, pilot, or operations.
- Outcome
- Risk
- Owner
- Next gate
Operations route
How the work will run
Read the workflow map, procedures, operating roles, metrics, first sprint, and buyer worksheet. This route shows whether staff can actually use, review, and improve the future process.
- Workflow
- Staff
- Support
- Improve
Technical and trust route
Where the boundaries live
Focus on records and work products, controls, risk assumptions, reference work products, trusted knowledge, runtime placement, and launch conditions before any private access expands.
- Source
- Access
- Runtime
- Rollback
Buyer session route
Turn reading into a working session
Use the discovery questions, role review route, buyer worksheet, and engagement fit ladder to prepare one process, one owner, one source map, and one next decision.
- Process
- Examples
- Questions
- Decision
Best use: read the route that matches your role, mark the questions that still need proof, and leave with one narrow decision instead of a vague AI wishlist.
Folium Systems Navigation map foliumsystems.com
01
Thesis
Folium turns AI urgency into operating capability.
AI access is becoming common, but practical AI operations are still hard. Folium Systems exists for the forward-engineering layer: the place where process design, software, data boundaries, controlled retrieval, agents, governance, staff adoption, and launch records have to work together before a business can depend on AI.
- Record
- Boundary
- Action
Investor premise
Capability is the scarce layer
Businesses can buy models and tools faster than they can redesign operations around them.
Folium role
Build the missing middle
Folium connects strategy, custom software, knowledge systems, agent controls, review rooms, and AI launch operations.
Why it matters
Review before dependency
The company packages inspectable records before asking buyers to trust private data, live systems, or production processes.
Claim discipline
Sourced market support only
Market size, adoption data, and financial assumptions belong in sourced diligence, not unsupported public claims.
Folium Systems Public-facing PDF foliumsystems.com
02
Deck standard
The deck follows the questions investors actually ask.
Strong investor decks are built around a fast diligence spine: purpose, problem, why now, customer, solution, product, market, model, traction signals, competition, team, financial plan, use of capital, and risk. Folium uses that structure while keeping public materials free of invented numbers.
- Decision grid
- Review lens
- Next step
| Investor question | Where Folium answers | Review standard |
|---|---|---|
| Why now? | AI pressure, workforce change, legacy-system drag, governance needs, and the shift from model access to operations. | Use sourced adoption and process-friction data before external distribution. |
| Who is the customer? | Mid-market operators, digital commerce teams, professional services, legacy operations, and regulated-adjacent processes. | Show target-account criteria, buyer personas, pain signals, and sales-cycle assumptions. |
| What is the product? | Controlled AI and software operations capability: process maps, workflow apps, trusted-data design, controlled retrieval, agents, private AI, launch rooms, and review files. | Show working examples, architecture, screenshots, delivery paths, and customer-approved records. |
| How does it make money? | Assessments, rapid build sprints, source-truth builds, agent implementations, private/hybrid AI work, and AI operations retainers. | Show pricing bands, delivery margin, repeatability, retention, and expansion only when approved. |
| Why Folium? | The digital manufacturing plant combines strategy, software, data, model orchestration, governance, and human adoption. | Test reusable assets, operating cadence, review quality, and implementation speed. |
| What does capital support? | Capacity, tooling, review quality, trust infrastructure, model/agent lab work, and go-to-market packaging. | Tie every dollar to milestones, runway, hiring, and measurable capability gains. |
Folium Systems Public-facing PDF foliumsystems.com
03
Problem
The buyer problem is wider than a chatbot, copilot, or automation recipe.
Customers are not only asking for AI. They are trying to modernize manual work, recover capacity, protect sensitive knowledge, connect legacy systems, train staff, and make decisions that leadership can defend.
- Checklist
- Owner path
- Release signal
- Knowledge is scattered across documents, tickets, inboxes, spreadsheets, product catalogs, policies, and staff memory.
- AI subscriptions can expand without clear data boundaries, source-of-truth rules, quality ownership, or rollback plans.
- Automation can move fragile processes faster without improving judgment, review, or accountability.
- Staff often receive AI pressure before they receive process redesign, training, or confidence-building support.
- Regulated-adjacent processes need records, human review, and compliance-aware launch discipline.
- Owners need an AI path that strengthens the business without forcing them to hire a full AI department first.
- Investor diligence should attach sourced buyer-adoption friction, implementation failure, and AI governance trend data.
Folium Systems Public-facing PDF foliumsystems.com
04
Market shift
The market is moving from model access to governed orchestration.
The next phase of AI is not one universal interface. It is multi-model, process-specific, source-grounded, private when needed, agent-assisted, measured, and operated through human-readable control systems.
- Decision grid
- Review lens
- Next step
| Shift | What changes | Folium response |
|---|---|---|
| Models become abundant | Strong AI services become available across many vendors and runtimes. | Folium focuses on implementation, fit, evaluation, and business process assembly. |
| Knowledge becomes strategic | Internal procedures, product data, customer context, and staff expertise become AI fuel. | Folium builds source boundaries, controlled retrieval, evaluation, and knowledge operating patterns. |
| Agents become operational | AI systems begin routing tasks, using tools, escalating, and assisting staff. | Folium designs permissions, blocked actions, human review, audit trails, and safe task lanes. |
| Private and hybrid AI matters | Some processes need local, private, portable, or cost-controlled runtime choices. | Folium maps cloud, local, private, and hybrid placement to business needs. |
| Governance moves earlier | AI risk cannot wait until after a working example has become dependency. | Folium uses review-before-production discipline, launch blockers, and known-limits records. |
Folium Systems Public-facing PDF foliumsystems.com
05
Solution
Folium is an AI operating-capability partner.
Folium helps businesses move from confusion to a working example, then from records to a controlled launch path. The service model blends AI consulting and forward engineering, custom application development, knowledge architecture, agent design, governance, and operational handoff.
- Record
- Boundary
- Action
Find the first process
Map the business process, pain, user roles, systems, data classes, edge cases, and first safe build.
Build the working example
Create clickable tools, portals, source-truth experiences, agent-assisted processes, dashboards, or integration surfaces.
Show behavior
Package screenshots, browser checks, evaluation cases, known limits, owner maps, and launch blockers.
Prepare operations
Define support, training, escalation, rollback, source maintenance, data handling, and improvement cadence.
Folium Systems Public-facing PDF foliumsystems.com
06
Forward engineering
Folium Forward Engineering is the method behind the operating layer.
Folium's category is not generic advisory and not model resale. Forward engineering is the disciplined delivery path from workflow discovery to production-ready operating capability.
- Decision grid
- Review lens
- Next step
| Delivery stage | Investor meaning | Customer output |
|---|---|---|
| Diagnose and scope | The company can qualify work by process pain, risk, data readiness, and first safe lane. | Embedded workflow review, technical scoping, and first-build decision. |
| Design and build | Reusable architecture, software surfaces, agent and trusted-data patterns, and integration modules can compound. | System design, working surface, integration build, and reviewable first build. |
| Evaluate and govern | Quality checks and controls make delivery more repeatable and safer to scale. | Evaluation harness, known-limit record, permissions, audit trail, and rollback triggers. |
| Launch and operate | Projects can turn into durable customer relationships instead of one-off walkthroughs. | Launch room, support plan, operating handoff, monitoring rhythm, and improvement backlog. |
This is why Folium can remain model-agnostic across hosted models, open models, local runtimes, controlled retrieval, agents, databases, legacy systems, and business workflows.
Folium Systems Public-facing PDF foliumsystems.com
07
Platform
The product/service platform is Folium's digital manufacturing plant.
Folium is built as a digital manufacturing plant: reusable services, build templates, agent patterns, model lanes, trusted-data patterns, governance records, and delivery rhythms that improve with each serious build.
- Decision grid
- Review lens
- Next step
| Plant layer | Reusable asset | Investor diligence lens |
|---|---|---|
| Intake and process | Discovery prompts, process maps, first-build selectors, buyer education tools. | How quickly can Folium identify a valuable, bounded first build? |
| Application build | Rapid app patterns, portals, dashboards, commerce processes, admin tools, browser verification. | How much implementation work becomes reusable? |
| Knowledge, source-of-truth, and data utilities | Source ingestion, retrieval boundaries, evaluation cases, memory rules, and plain-language explanation. | How does Folium preserve accuracy, custody, and usefulness? |
| Agents and controls | Task routing, tool permissions, human review, escalation, blocked-action logic, logs. | How does the company keep automation useful without overclaiming authority? |
| Governance and review files | Trust guides, risk registers, compliance-readiness reviews, known-limits records. | How does Folium make buyer confidence repeatable? |
| Operations | Support guides, training material, launch rooms, rollback paths, improvement loops. | How do records become durable customer capability? |
Folium Systems Public-facing PDF foliumsystems.com
08
Differentiation
Folium competes by assembling the operating system around AI.
The market sells pieces separately. Folium's differentiator is the ability to connect those pieces into a buyer-ready path: business process, custom software, controlled retrieval, agents, model placement, records, governance, compliance readiness, staff enablement, and AI operations.
- Record
- Boundary
- Action
Model capability becomes business capability
Folium does not need to out-model model providers. It turns model capability into business capability.
Advice turns into working records
The company builds working examples, tools, diagrams, review files, launch rooms, and operating records.
Runtime choices fit the work
Runtime decisions can include cloud, private, local, hybrid, open-source, and custom application paths.
Automation stays governed
Folium designs human review, source grounding, permission boundaries, and escalation around automated work.
Delivery includes the people who use it
Buyer language, staff training, adoption support, and workforce empowerment are part of the product.
Public examples open controlled diligence
Public-facing examples open the door; private diligence and customer-specific production work stay bounded.
Folium Systems Public-facing PDF foliumsystems.com
09
Traction signals
Current materials show capability signals.
This deck avoids revenue, customer-count, valuation, return, and pipeline claims. The public materials show that Folium can package a serious AI implementation story, build working digital examples, and separate public review from production dependency.
- Checklist
- Owner path
- Release signal
- Public website architecture with hub-based investor, review, trust, service, commerce, and resource paths.
- Downloadable public PDFs covering trust, security/procurement, AI risk, market positioning, and investor executive briefing.
- Review Vault patterns for sandbox review builds, rapid application builds, private knowledge assistant concepts, and advisor/copilot behavior.
- Digital manufacturing plant narrative with operating diagrams, launch reviews, and review-before-production framing.
- Browser-tested public experience across desktop, tablet, and mobile lanes including Brave verification in the current process.
- Controlled diligence should add approved review records, legally shareable customer records, pilot status, pipeline stage definitions, and dated logs.
Folium Systems Public-facing PDF foliumsystems.com
10
Business model
Folium can package AI capability through services, records, launches, and operations.
The current public deck frames model options without pricing claims or financial promises. Commercial detail belongs in controlled diligence after offer architecture, delivery capacity, legal structure, and customer pipeline are reviewed.
- Decision grid
- Review lens
- Next step
| Model component | What customer buys | Diligence detail to refine |
|---|---|---|
| Assessment and process discovery | A clear first AI process, risk map, data boundary, and test plan. | Pricing bands, conversion rate, delivery hours, qualification criteria. |
| Rapid application build | A working sandbox flow, browser-tested interface, review file, and next-step recommendation. | Delivery margin, reuse rate, scope controls, average cycle time. |
| Source-truth knowledge system build | Source-grounded assistant, retrieval controls, evaluation, and source maintenance plan. | Infrastructure cost, model choices, support model, customer data controls. |
| Agent and automation implementation | Permissioned agents, tool routing, human review, logs, and operating guides. | Risk tiering, support obligations, legal boundaries, audit records. |
| AI operations retainer | Monitoring, source refresh, quality review, staff support, improvement backlog. | Retention, staffing ratios, service-level boundaries, expansion triggers. |
| Private/local/hybrid AI build | Runtime placement, deployment architecture, data custody, portability planning. | Hardware/cloud economics, procurement path, support responsibility, security review. |
Folium Systems Public-facing PDF foliumsystems.com
11
Go to market
The first wedge is trust-led education for businesses that need AI but lack AI departments.
Folium's go-to-market should meet buyers where they are: overwhelmed by AI pressure, skeptical of hype, protective of staff knowledge, and looking for a practical first win. The public site, PDFs, resources, tools, and Review Vault are designed to shorten that education path.
- Record
- Boundary
- Action
Audience
Owners, operators, executives, and teams in SMB, digital commerce, professional services, legacy operations, workforce recovery, and regulated-adjacent processes.
Entry offer
A bounded process assessment or rapid build sprint that demonstrates value without touching live systems prematurely.
Buyer education
Plain-language pages, public PDFs, calculators, routers, checklists, diagrams, and review stories.
Conversion path
Move from public review to private discovery, scoped first build, controlled pilot, and AI operations only as records support it.
Sales support
Use buyer-specific talk tracks that translate AI, controlled retrieval, agents, governance, and private AI into business language.
Channel diligence
Investor review should test channel strategy, partner profile, target-account criteria, buyer personas, and market segment support.
Folium Systems Public-facing PDF foliumsystems.com
12
Moat
The moat is accumulated implementation knowledge turned into reusable systems.
Folium's defensibility should be tested through the depth of its delivery plant: reusable assets, review quality, customer-specific adaptation, internal tooling, staff enablement, source-grounded systems, governance discipline, and operating memory.
- Decision grid
- Review lens
- Next step
| Moat candidate | Why it can matter | How diligence should test it |
|---|---|---|
| Reusable review systems | Improves buyer confidence and shortens review cycles. | Inspect PDF generators, browser checks, templates, and review records. |
| Process and data patterns | Repeated business problems can be solved faster with known patterns. | Review examples across commerce, services, legacy ops, and regulated-adjacent work. |
| Agent governance patterns | Safe automation requires permissions, escalation, logs, and refusal boundaries. | Test blocked-action cases, review points, and failure handling. |
| Private/local/hybrid runtime expertise | Customers may need control over data, cost, latency, or vendor exposure. | Review deployment options, support limits, security posture, and portability. |
| Human adoption model | AI fails if staff reject it, misunderstand it, or cannot operate it. | Inspect training guides, role maps, support loops, and escalation design. |
| Operating cadence | AI systems need care after launch. | Review monitoring, source maintenance, improvement loops, and incident plans. |
Folium Systems Public-facing PDF foliumsystems.com
13
Roadmap
The roadmap should move from public review to repeatable operating capacity.
This deck avoids claiming completed capabilities beyond current public materials. The roadmap names logical workstreams to inspect in diligence and sequence through records rather than excitement.
- Decision grid
- Review lens
- Next step
| Horizon | Build focus | Review before expansion |
|---|---|---|
| Now | Strengthen public review material, investor room, service pages, Review Vault, trust guides, and buyer education. | Content review, browser checks, public-facing legal review, target buyer feedback. |
| Next | Package first repeatable offers for rapid application builds, source-truth readiness, commerce AI, and private AI assessment. | Scoped offer sheets, delivery checklists, reusable templates, acceptance criteria. |
| Then | Build deeper internal delivery tooling for process routing, record generation, evaluation, and launch rooms. | Tool walkthroughs, cycle-time records, quality checks, usage records, maintainability review. |
| Pilot | Run controlled customer pilots with clear data boundaries, human review, support, and known limits. | Pilot records, screenshots, eval cases, support notes, customer-approved records. |
| Operate | Move tested processes into monitored AI operations with source maintenance and improvement cadence. | Owner maps, incident paths, source freshness process, adoption metrics, renewal criteria. |
| Scale | Expand through industry lanes, partner channels, and stronger delivery capacity. | Sourced market support, channel economics, staffing plan, risk review, cash plan. |
Folium Systems Public-facing PDF foliumsystems.com
14
Team and operating model
Folium's operating model needs builders, translators, and record discipline.
The company should scale around a practical cross-functional pattern: understand the business, build the working example, govern the AI, support the staff, and document the record. Team details, founder history, advisors, hiring plan, and compensation belong in controlled diligence.
- Record
- Boundary
- Action
AI operating architect
Owns process design, runtime placement, data boundaries, first-build scope, and launch readiness.
Application builder
Creates the software surfaces: portals, dashboards, tools, APIs, integrations, and review experiences.
Source-truth and agent engineer
Builds source-grounded assistants, controlled retrieval where it fits, retrieval evaluation, task routing, permissions, and human review.
Governance and quality lead
Owns risk registers, test cases, known limits, compliance-readiness materials, and release records.
Customer translator
Turns fintech, AI, commerce, and operational complexity into buyer language and staff training.
Hiring sequence
Controlled diligence should add approved team bios, advisor roles, hiring order, operating cadence, and governance ownership.
Folium Systems Public-facing PDF foliumsystems.com
15
Use of capital
Capital should strengthen capacity, tooling, review quality, and trust infrastructure.
This page is public-facing and does not state an offering amount, investment terms, expected returns, valuation, revenue forecast, or legal solicitation. It names capability areas that can be refined after formal finance and legal review.
- Checklist
- Owner path
- Release signal
- Delivery capacity: expand the team and operating cadence required to run multiple scoped first builds without lowering quality.
- Internal tooling: build process assessment, source-truth readiness, review file, evaluation, browser checks, and launch-room tooling.
- Model and agent lab: improve prompt systems, agent patterns, evaluation harnesses, local/private runtime testing, and model comparison.
- Review portfolio: create public-facing and private diligence-ready examples across commerce, professional services, legacy operations, workforce recovery, and regulated-adjacent work.
- Trust infrastructure: strengthen security documentation, compliance-readiness review, data-boundary templates, support guides, and audit records.
- Go-to-market: package buyer education, partner material, sales enablement, working examples, case studies, and industry-specific offers.
- Formal diligence should add the capital amount, runway, hiring plan, milestone budget, financial model, and legal language only after approval.
Folium Systems Public-facing PDF foliumsystems.com
16
Risks and boundaries
Folium should be credible because it names what it will not overclaim.
Investor materials should preserve trust by separating review records from production, public copy from private diligence, AI support from regulated decision-making, and implementation capability from financial promises.
- Decision grid
- Review lens
- Next step
| Boundary | Public-facing position | Diligence follow-up |
|---|---|---|
| Financial claims | No revenue, return, valuation, margin, or customer-count claims are made in this deck. | Review approved financial statements, pipeline, assumptions, and legal offering documents. |
| Customer records | Public materials use working examples, PDFs, and qualitative capability signals only. | Share customer-approved records, contracts, references, or pilots through controlled diligence. |
| Regulated processes | AI can support review, routing, retrieval, explanation, and operations; it should not be framed as autonomous regulated decision authority. | Review legal boundaries, human review, compliance controls, and audit trails. |
| Data custody | Private, local, and hybrid AI are options to be designed by process and risk, not slogans. | Review architecture, security, retention, secrets handling, and vendor exposure. |
| Production readiness | A working example is not a production dependency until it passes launch readiness reviews. | Inspect acceptance criteria, support model, rollback plan, monitoring, and owner map. |
| Scalability | Repeatability is a thesis to test through tooling, process, and records. | Test delivery throughput, staffing plan, quality controls, and reusable asset maturity. |
Folium Systems Public-facing PDF foliumsystems.com
17
Next diligence step
The next investor conversation should inspect the engine.
The right next step is not a bigger promise. It is a controlled review of the assets, operating model, commercial assumptions, technical architecture, and first repeatable customer wedge.
- Checklist
- Owner path
- Release signal
- Review the public Review Vault, investor executive brief, market positioning brief, trust guide, AI risk launch standard, and security/procurement review.
- Walk through one rapid application build and one trusted-data or agent process from discovery to launch-readiness records.
- Identify the first target customer segment and define the narrowest paid first-build offer that demonstrates repeatable value.
- Inspect the digital manufacturing plant roadmap: reusable modules, internal tooling, evaluation harnesses, record generators, and launch rooms.
- Review legal and finance materials separately before any offering, terms, forecasts, or investor-specific commitments are discussed.
- Attach sourced market notes, approved diligence exhibits, customer-approved records, and finance materials before external investor distribution.
This pitch deck is a public-facing review packet. It is not an offer to sell securities, investment advice, a forecast, or a guarantee of financial performance.
Folium Systems Public-facing PDF foliumsystems.com
18
Pitch appendix
The pitch deck should now reference the expanded diligence shelf.
The pitch deck remains the boardroom spine. The expanded PDFs give each major claim a deeper public companion: category, services, why us, differentiation, plant, audit, staff, local AI, commerce, and forward engineering for investors.
- Decision grid
- Review lens
- Next step
| Deck claim | Supporting packet | Diligence use |
|---|---|---|
| Why now | Why Folium / Five Ws | Clarifies buyer pressure and timing. |
| Solution | What Folium Does | Explains the service architecture. |
| Method | Forward Engineering Field Guide | Details the delivery discipline. |
| Moat | Digital Manufacturing Plant Brief | Shows compounding delivery assets. |
| Market position | Folium Differentiation Brief | Compares against other AI categories. |
| Expansion | Local AI and Commerce packets | Shows high-value vertical paths. |
Folium Systems Public-facing PDF foliumsystems.com
19
Reader route
Use the packet by role, not only from front to back.
The strongest review happens when each stakeholder reads the pages that match their decision rights. This route helps a buyer turn the packet into a working session instead of a passive download.
- Decision grid
- Review lens
- Next step
| Reviewer | What to inspect | Question to answer |
|---|---|---|
| Owner or CEO | Value, risk, first process, launch gates, and next-stage decision. | Is this a controlled way to move from AI pressure to capability? |
| Operations lead | Workflow steps, people affected, support path, and improvement rhythm. | Can the team operate this without creating a new hidden burden? |
| Technical lead | Systems, runtime, integrations, logs, fallback, and data boundaries. | Can the architecture be supported and secured? |
| Security or procurement | Access, retention, provider exposure, blocked data, permissions, and rollback. | What must be true before private access expands? |
| Staff manager | Training, role clarity, human review, correction path, and adoption risk. | Will this strengthen the people doing the work? |
| Investor or partner | Category, repeatability, public boundary, and diligence path. | What deeper records should be requested before believing the thesis? |
Folium Systems Public-facing PDF foliumsystems.com
20
Working-session worksheet
Bring these answers into the next Folium conversation.
A printable PDF should help the buyer prepare. These prompts keep the conversation attached to real work, real systems, real people, and an honest boundary between public review and private implementation.
- Checklist
- Owner path
- Release signal
- Name the one workflow that hurts most today and the person who owns it.
- List every system, file, inbox, store, database, spreadsheet, vendor, or manual handoff the workflow touches.
- Separate data into public, internal, customer, regulated, confidential, credential, and blocked classes.
- Identify which steps are slow, duplicated, risky, customer-visible, staff-heavy, or expensive.
- Write down what AI may draft, retrieve, recommend, route, block, or escalate.
- Write down what AI must not execute without human approval.
- Bring examples of good output, bad output, common exceptions, missing data, and escalation moments.
- Decide what record would justify the next step: audit, first build, architecture review, pilot, or operations.
Folium Systems Public-facing PDF foliumsystems.com
21
Decision matrix
The next step should be earned by the record.
Folium's public packets are built to create a practical decision, not only a favorable impression. Use this matrix to choose the next move after review.
- Decision grid
- Review lens
- Next step
| Decision | Use when | Expected next record |
|---|---|---|
| Stop | The process has no owner, no clear value, or unsafe data pressure. | Stop note and conditions that would need to change. |
| Refine | The pain is real but the workflow, source truth, or approval path is unclear. | Revised process map and missing-information list. |
| Audit | The buyer sees the need but does not know which AI lane should come first. | AI systems audit, inventory, scorecard, and first-lane recommendation. |
| First build | One safe process, owner, source boundary, and desired output are clear. | Working surface, known limits, browser checks, and next-stage blockers. |
| Architecture review | A useful build exists but private access, runtime, support, or authority needs review. | Data boundary, runtime matrix, permission map, and rollback path. |
| Operate | A pilot has value, owners, support, monitoring, and improvement rhythm. | AI operations cadence, source refresh plan, release notes, and issue loop. |
Folium Systems Public-facing PDF foliumsystems.com
22
Plain-language glossary
The buyer should not need to speak engineer to read the packet.
Folium uses technical terms when needed, but a public packet should translate them into operating language. The goal is to help the buyer understand the decision, not admire the vocabulary.
- Decision grid
- Review lens
- Next step
| Term | Plain meaning | Why it matters |
|---|---|---|
| RAG | AI answers from approved company material instead of memory alone. | It keeps answers tied to business sources. |
| Agent | A guided AI worker that can follow a task path with tools and limits. | It needs permission, logging, and human review. |
| Runtime | Where the AI work runs: cloud, private endpoint, local machine, or hybrid path. | It affects privacy, cost, speed, control, and support. |
| Evaluation | A test set that checks whether the system behaves correctly on real tasks. | It exposes failures before the business depends on the system. |
| Governance | The rules for data, access, authority, logs, review, rollback, and ownership. | It keeps AI useful without giving it unmanaged power. |
| Launch room | The operating board for owners, support, blockers, training, incidents, and next releases. | It turns a build into a system the business can run. |
Folium Systems Public-facing PDF foliumsystems.com
23
Next step
The pitch is strongest when it stays record-led.
Use this deck to start the investor conversation, then move into controlled diligence for financials, customer records, legal materials, proprietary tooling, and technical architecture.
Bring the process
Name the business process, the systems involved, the people affected, and the decision this PDF should support.
Separate review from production
Keep public examples, sandbox review, pilot access, and production dependency in separate stages with clear owners.
Ask for the record
Request screenshots, browser checks, known limits, launch blockers, support plans, and the next approval path.
Folium Systems Public-facing PDF foliumsystems.com