Skip to content

Services
AI Solutions
Industries
Hire Developers
Company
Resources
Discuss Your Project

Discuss Your Project

Home / Insights / Topics / Generative AI and LLMs

Generative AI and LLMs

Practical guides to adding large language models to products: retrieval, fine-tuning, chatbots and internal knowledge assistants.

About this topic

Large language models are now easy to call and hard to run well. This topic covers the design choices that decide whether an LLM feature is useful in production: when retrieval-augmented generation is enough, when fine-tuning helps, how to keep answers tied to approved sources and how to add generative features to an application that already has users.

Each guide is written for teams that have to live with the result. We cover permissions, evaluation sets, citation checks and fallback behaviour alongside the user-facing feature, because those are the parts that usually decide cost, risk and adoption.

Start here

LLMGenerative AI and LLMsRAG or fine-tuning? Giving an AI model your company knowledgeWhen to use retrieval-augmented generation, when fine-tuning helps and how to combine them to build AI features grounded in your own data.3 min read

31 articles

All Generative AI and LLMs articles

LLMGenerative AI and LLMsAI in Accounting Software: Use Cases, Model Choices and the Controls to Build FirstWhere AI helps in accounting software — capture, coding, matching, reconciliation, anomaly detection and reporting — which techniques suit each task, a worked invoice example, and the controls auditors expect.9 min read

LLMGenerative AI and LLMsAI in Electronic Health Records: Practical Use Cases, Integration and SafeguardsHow AI is being added to electronic health record workflows — documentation, inbox management, coding, summaries, search and alerts — with integration patterns, evaluation, governance and privacy safeguards.4 min read

LLMGenerative AI and LLMsBuilding HIPAA-Compliant AI: Architecture, Data Handling and Vendor ChoicesHow to design AI and LLM features that handle protected health information under HIPAA: where PHI flows, architecture patterns, de-identification, vendor agreements, logging, breach readiness and the Security Rule update.6 min read

LLMGenerative AI and LLMsHow to Build an Internal AI Knowledge Assistant for Your CompanyPlan an internal AI knowledge assistant with approved sources, permission-aware retrieval, source references and an evaluation process for users.5 min read

LLMGenerative AI and LLMsReducing LLM Token and Cloud Costs without Losing Answer QualityA practical guide to cutting LLM token and generative AI cloud costs: where tokens leak, nine techniques with trade-offs, a benchmark plan and safe rollout.12 min read

LLMGenerative AI and LLMsRAG in Healthcare: Use Cases, Architecture and Safeguards for Grounded AI AssistantsHow retrieval-augmented generation works in healthcare: suitable and unsuitable use cases, a reference architecture, privacy and regulatory considerations, evaluation and a phased rollout plan.11 min read

LLMGenerative AI and LLMsTransformers and Language Models Explained for Product TeamsA plain-language explanation of transformer models and language models: attention, tokens, context windows, encoder and decoder models, pre-training and fine-tuning, strengths, limits and what they mean for product decisions.3 min read

LLMGenerative AI and LLMsVision Language Models for Business: How They Work, Where They Help and How to Evaluate ThemWhat vision language models are, how they combine image and text understanding, business use cases, open versus hosted options, failure modes such as visual hallucination, and how to evaluate them on your own images.6 min read

LLMGenerative AI and LLMsHow to Add AI Features to an Existing Web or Mobile ApplicationAdd AI to an existing web or mobile application with clear integration boundaries, evaluations, fallback behavior and an incremental release plan.5 min read

LLMGenerative AI and LLMsCustom AI Chatbot vs Off the Shelf Tools: Which Should Your Business Choose?Compare custom AI chatbots with off-the-shelf tools by workflow fit, integrations, data controls and total ownership cost before deciding to build.5 min read

LLMGenerative AI and LLMsRAG or fine-tuning? Giving an AI model your company knowledgeWhen to use retrieval-augmented generation, when fine-tuning helps and how to combine them to build AI features grounded in your own data.3 min read

AGAI Agents and AutomationChoosing an Agentic AI Framework: LangGraph, CrewAI, OpenAI Agents SDK, Google ADK and Microsoft Agent FrameworkA practical comparison of agentic AI frameworks — LangGraph, CrewAI, OpenAI Agents SDK, Google ADK, Microsoft Agent Framework and others — with selection criteria and a scored shortlist.10 min read

AGAI Agents and AutomationAgentic AI vs Generative AI: A Practical Decision GuideUnderstand the real difference between generative and agentic AI, when each fits, the workflow patterns in between and the controls agents need before they take actions.7 min read

ARArchitecture and ModernizationChoosing AI Application Architecture: Six Patterns and Their Trade-offsCompare six AI application architecture patterns on one running example, with a decision matrix, trust boundaries, failure handling, evaluation and an ADR template.19 min read

$Software Costs and BudgetsAI FinOps for Product Teams: Measuring Cost per Successful WorkflowA practical AI FinOps guide for product teams: cost per successful workflow, a worked token and review-cost model, attribution, budgets and safe optimisation.12 min read

AIAI and Machine LearningAI in Banking Software: Use Cases, Risks and How to Implement ResponsiblyPractical AI use cases in banking — onboarding, fraud and AML, credit, customer service, operations and compliance — with the model risk, explainability, data protection and governance requirements that apply.3 min read

AIAI and Machine LearningAI in Medical Image Analysis: What It Does Well, Where It Falls Short and How to Deploy ItHow AI supports medical imaging — detection, triage, measurement and workflow — the regulatory picture, evidence and validation, integration with PACS and RIS, limitations and how to deploy responsibly.3 min read

AIAI and Machine LearningAI, Machine Learning, Deep Learning and Generative AI: What’s the Difference and Which Do You Need?Clear definitions of artificial intelligence, machine learning, deep learning and generative AI, how they relate, what each needs in data and cost, and how to choose the right approach for a business problem.3 min read

ARArchitecture and ModernizationBuilding APIs for AI Features: Streaming, Limits, Versioning and SafetyHow to design the API between your apps and the AI features behind them: streaming, long-running jobs, token budgets, rate limits, error handling, versioning, security and observability.9 min read

AGAI Agents and AutomationContext Engineering for AI Agents: Giving Models the Right Information at the Right TimeWhat context engineering is, why agents fail without it, the techniques that work — retrieval, compaction, notes, sub-agents and tool design — and how to measure and secure an agent’s context.6 min read

PDProduct Design and DiscoveryConversational UI: When Chat Is the Right Interface and How to Design It WellWhat conversational interfaces are, when chat or voice beats forms and menus, design principles for LLM-powered assistants, hybrid patterns with buttons and cards, accessibility, failure handling and metrics.5 min read

AGAI Agents and AutomationCustomer Support AI Agents: A Reference Architecture with Escalation and GuardrailsA reference architecture for customer support AI agents: channels, orchestration, retrieval, action tools with permissions, escalation to humans, evaluation, metrics and rollout.6 min read

AIAI and Machine LearningA Data Strategy for AI: What to Decide Before You BuildHow to build a data strategy for AI: start from use cases, inventory and assess data, decide ownership and governance, design pipelines for training, retrieval and evaluation, and measure data readiness — with a worked example and 90-day plan.6 min read

AIAI and Machine LearningBlack Box AI and Explainability: What Product Teams Need to KnowWhat makes AI a “black box”, why it matters for trust, regulation and debugging, the main explainability techniques such as SHAP, LIME and counterfactuals, explaining LLM outputs, and how to choose the right level of transparency.4 min read

AIAI and Machine LearningMLOps and LLMOps: Running Machine Learning and AI Features Reliably in ProductionWhat MLOps is, how it differs from DevOps, the lifecycle from data to monitoring, how LLMOps adapts it for generative AI, a maturity model and the practices that keep AI features reliable.3 min read

AIAI and Machine LearningResponsible AI in Practice: Turning Principles into Product DecisionsHow product teams turn responsible AI principles into practice: risk classification, the NIST AI RMF, ISO/IEC 42001, the EU AI Act timeline, bias testing, transparency, human oversight and governance for generative AI and agents.4 min read

AIAI and Machine LearningSovereign AI: What It Means for Businesses Choosing Where Their AI RunsWhat sovereign AI means at national and company level, why data residency, control and resilience matter, deployment options from regional cloud to self-hosted open models, trade-offs and a decision framework.4 min read

AIAI and Machine LearningSpeech-to-Text in Products: How Modern ASR Works and How to Use It WellHow automatic speech recognition works today, how to measure accuracy with word error rate, choosing between APIs and self-hosted models, handling accents, noise and privacy, and common product use cases.3 min read

AGAI Agents and AutomationWhat Are AI Agents? A Practical Guide for Product and Engineering TeamsWhat AI agents are, how the agent loop works, the five components every agent needs, levels of autonomy, where agents fit, and the controls to put in place first.9 min read

$Software Costs and BudgetsAI Chatbot Development Cost: Setup, Integration and Ongoing ExpensesUnderstand AI chatbot development costs across setup, integrations, model usage and maintenance, with a practical way to estimate recurring expenses.5 min read

AGAI Agents and AutomationAI Document Processing: Use Cases, Costs and Human ReviewExplore AI document processing through intake, extraction, validation and human review, with a practical evaluation and cost model for one workflow.6 min read

Related topics

AI and Machine Learning AI Agents and Automation Architecture and Modernization Business Operations and Integration Software Costs and Budgets All topics →

Planning work in generative ai and llms?

Tell us about your product, users and constraints. We will suggest a sensible first step.

Explore LLM Applications

Discuss Your Project

AI & software engineering from Hanoi, Vietnam. Web and mobile products, practical AI integration and engineering teams for international companies.

★★★★★ 5.0 · 57 reviews

LinkedIn Facebook Clutch

Services

AI Development
AI Agents
Mobile Apps
Web & SaaS
Cloud & DevOps
QA & Testing

Solutions

AI Automation
Document AI
Knowledge Search
Dedicated Teams
Hire Developers
Industries

Company

About
Our Work
Insights
How We Work
Engineering
Contact

Contact

contact@politetechsoftware.com

3rd Floor, Trung Yen Plaza, 1 Trung Hoa Street, Yen Hoa Ward, Hanoi, Vietnam

© 2026 PoliteTech Software. All rights reserved. · Privacy Policy · Terms · FAQs · Cookie settings