docs: make organisation profile accurate and verifiable

.github/workflows/profile-quality.yml

@@ -0,0 +1,22 @@
+name: Profile quality
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+permissions:
+  contents: read
+
+jobs:
+  markdown:
+    runs-on: ubuntu-24.04
+    steps:
+      - uses: actions/checkout@v4
+      - uses: DavidAnson/markdownlint-cli2-action@v20
+        with:
+          globs: |
+            README.md
+            profile/*.md
+          config: .markdownlint.json

.markdownlint.json

@@ -0,0 +1,4 @@
+{
+  "default": true,
+  "MD013": false
+}

README.md

@@ -1,90 +1,17 @@
-<div align="center">
-  <img src="https://raw.githubusercontent.com/AcadifySolution/.github/main/profile/assets/profile-animation-root.svg" width="100%" alt="Acadify AI Labs Brand Header">
-</div>
+# Acadify Solution organisation profile
 
-<br>
+The public organisation landing page is maintained in
+[profile/README.md](profile/README.md) and displayed at
+[github.com/AcadifySolution](https://github.com/AcadifySolution).
 
-<div align="center">
-  <p>
-    <a href="https://acadifysolution.com">Website</a> &nbsp;•&nbsp;
-    <a href="mailto:contact@acadifysolution.com">Email Inquiries</a> &nbsp;•&nbsp;
-    <a href="https://calendly.com/acadify-online/30min">Book Technical Consultation</a> &nbsp;•&nbsp;
-    <a href="https://github.com/AcadifySolution">Main Organization</a>
-  </p>
-
-  <p>
-    <img src="https://img.shields.io/badge/Services-Model_Training_&_Evaluation-cf6b48?style=flat-square" alt="Services">
-    <img src="https://img.shields.io/badge/Lab-Silicon_Valley_%26_India-0d3169?style=flat-square" alt="Location">
-    <img src="https://img.shields.io/badge/Security-VPC_Deployment_Audits-10b981?style=flat-square" alt="Security Scope">
-  </p>
-</div>
+## Maintaining the profile
 
----
+- Describe capabilities supported by the linked repository source.
+- Keep project links under the organisation and verify they remain public.
+- Distinguish examples, mock behavior, and deployment scaffolding from operating services.
+- Do not publish client information, personal account details, invented metrics,
+  certifications, office locations, or unverified security guarantees.
+- Propose focused changes through a branch and pull request.
 
-### 🔬 Acadify AI Labs
-
-Acadify AI Labs is the model testing, training, and evaluation division of Acadify Solution. We specialize in LLM behavioral analysis, bias diagnostics, adversarial safety auditing, and Supervised Fine-Tuning (SFT) dataset validation. We design evaluation metrics to help engineering teams verify model accuracy, consistency, and compliance before promoting systems to production.
-
----
-
-### ⚙️ The AI System Review (ASR) Methodology
-
-We evaluate production readiness through a structured 3-phase review process:
-
-1. **Context Engineering:** Mapping target user journeys and systems architecture to define high-impact evaluation datasets that mirror live production usage.
-2. **Pressure Simulation:** Simulating sustained interaction sequences (long sessions and complex prompt sequences) to analyze model reasoning preservation and logic drift over time.
-3. **Failure Mode Analysis:** Exposing latent vulnerabilities, hallucinations, and safety compliance failures that standard unit testing and static benchmarks do not capture.
-
----
-
-### 💻 Code Sandbox: Automated Semantic Faithfulness Evaluation
-
-The following Python snippet demonstrates how we run programmatic semantic checks on RAG pipeline outputs using a custom evaluation class integrated into CI test runners:
-
-```python
-import pytest
-from ai_evals import Evaluator, metrics
-
-# Evaluates that candidate outputs remain faithful to the retrieved context
-def test_rag_faithfulness_metric():
-    evaluator = Evaluator(model="claude-3-5-sonnet-20241022")
-
-    query = "What is the recorded API gateway latency?"
-    context = "The API gateway logs report an average processing latency of 24ms."
-    candidate_output = "The gateway processes requests with an average latency of 24ms."
-
-    # Calculates a semantic overlap and faithfulness score between 0.0 and 1.0
-    score = evaluator.score(
-        query=query,
-        context=context,
-        output=candidate_output,
-        metric=metrics.FAITHFULNESS
-    )
-
-    # Assert output matches the context above a semantic threshold of 0.85
-    assert score >= 0.85
-```
-
----
-
-### 🛡️ Active Testing &amp; Diagnostics Repositories
-We maintain reference architectures and libraries for evaluating model performance and securing LLM integrations:
-
-*   **[`ai-evals-framework`](https://github.com/AcadifySolution/ai-evals-framework)** — Automated quality assurance frameworks and semantic similarity checks.
-*   **[`llm-security-proxy`](https://github.com/AcadifySolution/llm-security-proxy)** — Real-time PII redaction and prompt injection classification gateways.
-*   **[`fine-tuning-pipeline`](https://github.com/AcadifySolution/fine-tuning-pipeline)** — Modular training configurations for LoRA/QLoRA model adaptation.
-*   **[`asr-feedback`](https://github.com/AcadifySolution/asr-feedback)** — Structured frameworks to capture model performance drift and diagnostics feedback.
-
----
-
-### 📬 Technical Consultations
-
-For questions regarding evaluation protocols, VPC deployment architectures, or custom dataset generation models, schedule a call with one of our engineers:
-
-<div align="center">
-  <br>
-  <a href="https://calendly.com/acadify-online/30min">Schedule Technical Call</a> &nbsp;•&nbsp;
-  <a href="mailto:contact@acadifysolution.com">contact@acadifysolution.com</a>
-  <br><br>
-  <sub>&copy; 2026 Acadify Solution. All rights reserved. Managed by @acadify-admin.</sub>
-</div>
+Profile content does not establish shared repository policies. Refer to each
+project's README, contribution guide, security policy, and license for its terms.

profile/ORG_PROFILE.md

@@ -1,25 +1,4 @@
-# Acadify Solution — Organization Profile
+# Organisation profile source
 
-Acadify Solution is an AI engineering company focused on production LLM systems.
-
-We build and deploy:
-- LLM evaluation and AI quality systems
-- Fine-tuning and model adaptation pipelines
-- Retrieval-augmented generation (RAG)
-- Autonomous AI agent workflows
-- LLM security gateways and PII protection
-- Cloud AI infrastructure and deployment
-
-Primary website: https://acadifysolution.com
-Organization profile: https://github.com/AcadifySolution
-
-Public engineering projects:
-- https://github.com/AcadifySolution/ai-evals-framework
-- https://github.com/AcadifySolution/llm-security-proxy
-- https://github.com/AcadifySolution/fine-tuning-pipeline
-- https://github.com/AcadifySolution/enterprise-rag-boilerplate
-- https://github.com/AcadifySolution/autonomous-agent-swarm
-- https://github.com/AcadifySolution/asr-feedback
-
-For AI engineering and implementation inquiries:
-contact@acadifysolution.com
+The maintained organisation profile is [README.md](README.md).
+Update that file through a pull request to keep the GitHub landing page current.

profile/README.md

@@ -1,110 +1,47 @@
 # Acadify Solution
 
-**Enterprise AI engineering, model development, evaluation, and deployment.**
-
-Acadify Solution builds practical AI systems for companies that need to move from prototype to production. Our public engineering work covers **LLM evaluation, fine-tuning, RAG systems, autonomous agents, AI security, and deployment infrastructure**.
-
-**Website:** https://acadifysolution.com  
-**AI Platform & Labs:** https://ai.acadifysolution.com/  
-**Engineering & AI Blog:** https://acadifysolution.com/blogs  
-**GitHub:** https://github.com/AcadifySolution  
-**LinkedIn:** https://www.linkedin.com/company/acadify-solutions/  
-**General:** hello@acadifysolution.com
-
-## What we build
-
-| Area | Focus |
-|---|---|
-| LLM Engineering | Fine-tuning, LoRA/QLoRA, model adaptation and deployment |
-| AI Evaluation | Accuracy, semantic similarity, grounding, safety, regression testing |
-| RAG | Retrieval pipelines, document processing, vector search and reranking |
-| AI Agents | Tool use, orchestration, validation and reliable workflow execution |
-| AI Security | LLM gateways, PII protection, prompt-risk controls and auditability |
-| Cloud AI | AWS, GCP, Docker, Kubernetes and Terraform-based infrastructure |
-
-## Public engineering projects
-
-- **AI Evals Framework** — repeatable evaluation of LLM and RAG outputs with correctness, semantic, grounding and safety metrics.  
-  https://github.com/AcadifySolution/ai-evals-framework
-
-- **LLM Security Proxy** — security gateway for LLM traffic with request controls, PII handling, guardrails and audit logging.  
-  https://github.com/AcadifySolution/llm-security-proxy
-
-- **Fine-Tuning Pipeline** — LoRA/QLoRA SFT pipeline with Accelerate, DeepSpeed and SageMaker deployment scaffolding.  
-  https://github.com/AcadifySolution/fine-tuning-pipeline
-
-- **Enterprise RAG Boilerplate** — production-oriented RAG reference architecture and implementation patterns.  
-  https://github.com/AcadifySolution/enterprise-rag-boilerplate
-
-- **Autonomous Agent Swarm** — orchestration patterns for multi-agent workflows, retries, checkpoints and operational controls.  
-  https://github.com/AcadifySolution/autonomous-agent-swarm
-
-- **ASR Feedback** — structured evaluation and feedback infrastructure for AI systems.  
-  https://github.com/AcadifySolution/asr-feedback
-
-## Official resources
-
-### AI Platform & Labs
-
-**https://ai.acadifysolution.com/**
-
-Our AI platform and lab environment for LLM evaluation, RAG, AI architecture reviews, reliability evaluation and related platform work.
-
-**AI Platform & Labs:** ai@acadifysolution.com
-
-### Engineering & AI Blog
-
-**https://acadifysolution.com/blogs**
-
-Articles and technical content covering AI engineering, LLMs, RAG, evaluation, security, development and practical implementation.
-
-## Business & engagement
-
-### Enterprise Sales
-Proposals, scoping, quotes, custom statements of work, dedicated engineering pods and enterprise MVP delivery.
-
-**sales@acadifysolution.com**
-
-### Strategic Partnerships
-Technology alliances, cloud reseller partnerships and executive co-development engagements.
-
-**business@acadifysolution.com**
-
-### AI Platform & Labs
-LLM evaluation, RAG, platform sandbox, architecture reviews and reliability evaluations.
-
-**ai@acadifysolution.com**
-
-### Careers & Talent
-Engineering pods, AI engineers, full-stack architects and DevSecOps opportunities.
-
-**jobs@acadifysolution.com**
-
-### General Communications
-General inquiries, event invitations, press outreach and technical queries.
-
-**hello@acadifysolution.com**
-
-## Our engineering approach
-
-We focus on systems that are:
-
-**Reliable** — deterministic checks, validation, failure handling and regression coverage.  
-**Observable** — useful telemetry, evaluation reports and operational signals.  
-**Secure** — least-privilege infrastructure, secrets hygiene and protection of sensitive model data.  
-**Practical** — architecture chosen around the actual product, workload and deployment constraints.
-
-## For developers, AI teams and enterprises
-
-Explore the public repositories for reusable reference implementations, evaluation patterns, infrastructure examples and engineering practices.
-
-For implementation, architecture, AI evaluation, model pipelines or deployment work:
-
-**sales@acadifysolution.com**  
-https://acadifysolution.com
-
-> Public repositories describe representative engineering work. Specific client engagements, metrics, infrastructure and security controls may vary by project and are not disclosed unless authorized.
-
-## Keywords
-
-Acadify Solution, Acadify AI, enterprise AI, AI engineering, LLM engineering, generative AI, large language models, LLM evaluation, AI evaluation, model evaluation, RAG, retrieval augmented generation, LLM fine-tuning, LoRA, QLoRA, autonomous agents, AI agents, LLM security, AI security, PII protection, prompt injection defense, AI infrastructure, AWS SageMaker, Kubernetes, Terraform, machine learning engineering, AI development company, AI solutions, AI platform, AI labs, AI R&D, enterprise MVP development, AI architecture, AI reliability.
+AI engineering, evaluation, and practical software development.
+
+We build tools and reference implementations for teams working with language models,
+retrieval, model adaptation, and application delivery. Start with the project that
+matches your use case, then follow its setup guide and tests.
+
+[Website](https://acadifysolution.com) ·
+[Acadify AI Lab](https://ai.acadifysolution.com/) ·
+[Contact](mailto:contact@acadifysolution.com)
+
+## Explore the engineering work
+
+| Project | What you can explore |
+| --- | --- |
+| [AI Evals Framework](https://github.com/AcadifySolution/ai-evals-framework) | Dataset-based evaluation, metric thresholds, and ordered evaluation reports |
+| [LLM Security Proxy](https://github.com/AcadifySolution/llm-security-proxy) | API-key access, pattern-based prompt screening, PII redaction, and audit events |
+| [Fine-Tuning Pipeline](https://github.com/AcadifySolution/fine-tuning-pipeline) | Dataset preparation, LoRA/QLoRA training code, and training configuration examples |
+| [Enterprise RAG Boilerplate](https://github.com/AcadifySolution/enterprise-rag-boilerplate) | Document ingestion, metadata-filtered retrieval, vector-store adapters, and response checks |
+| [Autonomous Agent Swarm](https://github.com/AcadifySolution/autonomous-agent-swarm) | Task queues, workflow dependencies, persisted state, recovery, and human approval gates |
+| [Aegis MVP Launchpad](https://github.com/AcadifySolution/aegis-mvp-launchpad) | A deterministic planning demo with a React dashboard and FastAPI backend |
+| [ASR Feedback](https://github.com/AcadifySolution/asr-feedback) | Feedback schemas, example records, SDK source, and evaluation playbooks |
+| [Sales Automation](https://github.com/AcadifySolution/Sales-Automation) | A local sales dashboard and configurable lead, CRM, and follow-up workflows |
+| [Engineering Standards](https://github.com/AcadifySolution/engineering-standards) | Git branching, review practices, coding guidance, and release checklists |
+
+## Evaluate before adopting
+
+These repositories are reference implementations and development tools. Read each
+README and security policy for implemented behavior, mock modes, configuration,
+and limitations. Passing unit tests does not establish production readiness,
+regulatory compliance, or complete protection from prompt injection.
+
+Check the repository's **Actions** tab for current validation results and its
+**Pull requests** tab for proposed changes. Infrastructure examples require a
+separate review of access controls, costs, networking, and operational needs
+before deployment. Project-specific license terms are in each repository.
+
+## Work with us or contribute
+
+For implementation and architecture inquiries, contact
+[contact@acadifysolution.com](mailto:contact@acadifysolution.com).
+
+For code contributions, follow the target repository's `CONTRIBUTING.md`, open a
+focused branch and pull request, and include the checks you ran. For suspected
+security issues, use that repository's `SECURITY.md` reporting instructions and
+keep credentials and sensitive records out of public issues.