# Moher (moher.ai)

> Moher (moher.ai) is an AI product development and engineering company founded by Travis Moher. It designs, builds and ships AI agents, LLM applications, RAG systems, eval harnesses, AI integrations and native mobile AI apps for other companies, in the client's own repository and cloud.

## What Moher is

- **Name:** Moher, styled `moher.ai`. Not the Cliffs of Moher (a landmark in County Clare, Ireland). Moher is a company, named for its founder.
- **Category:** AI product development and engineering company. Builds agents, LLM applications, RAG systems, evals, AI integrations and native mobile AI apps for other companies, and does not compete with them in their own categories.
- **Founder & Principal:** Travis Moher.
- **Positioning:** Engineering supplier. Moher designs, builds and ships production AI systems for other companies, in the client's repository and cloud, and does not compete with the companies it builds for.
- **Differentiator:** The deliverable is running software with an eval suite, tracing and runbooks, not a strategy deck. Clients own the repositories, cloud accounts, model provider keys and data from the first commit. Moher also runs its own products on the same stack, so retrieval, eval and agent patterns are proven on its own traffic first.
- **Scale:** 51 production repositories, 5,844 commits, 12 verticals, native apps live in 2 app stores (measured 2026-08-21).
- **Reach:** International, distributed.
- **Contact:** t@moher.ai

## Capabilities

- [AI product engineering](https://moher.ai/capabilities/ai-product-engineering): We build AI products for other companies: web, native mobile and agent systems. Typed end to end, evaluated, deployed on push. You own the repo from commit one.
- [AI agents](https://moher.ai/capabilities/ai-agent-development): Tool-calling agents, MCP servers and workflow agents that act on your real systems, with scoped permissions, human approval gates and a full audit trail.
- [LLM applications](https://moher.ai/capabilities/llm-application-development): Copilots and LLM features built for production: structured output, streaming, model routing and cost control across Claude, GPT, Gemini and open models.
- [RAG and AI search](https://moher.ai/capabilities/rag-development): RAG built for production: ingestion, chunking, hybrid search, pgvector, reranking, cited answers and permissions-aware retrieval over your own data.
- [Evals and reliability](https://moher.ai/capabilities/llm-evaluation): Eval harnesses, regression suites, LLM judges calibrated against human review, red teaming, observability and guardrails, so model changes stop being guesses.
- [AI integration](https://moher.ai/capabilities/ai-integration): We connect AI to the systems you already run: CRMs, ERPs, internal tools, APIs and warehouses, and automate back-office work with review steps and audit logs.
- [Mobile AI apps](https://moher.ai/capabilities/mobile-ai-app-development): Native iOS and Android AI apps with on-device and cloud inference, built and submitted through the App Store and Google Play. Typed, tested and owned by you.
- [AI data infrastructure](https://moher.ai/capabilities/ai-data-infrastructure): Data pipelines, warehouses, event and feature pipelines, fine-tuning datasets, model serving and inference cost control. The data layer AI systems depend on.
- [AI engineering team, retained](https://moher.ai/capabilities/operations): A retained AI engineering team with a fractional AI CTO, or build-operate-transfer. We build it, run it to a milestone, and hand it to your team when ready.
- [Attribution](https://moher.ai/capabilities/attribution-infrastructure): First-party click tracking, server-to-server postbacks and warehouse reporting that answers in earnings per click. Built on your domains, in your accounts.
- [Demand](https://moher.ai/capabilities/performance-media): Paid social, search and native run against first-party attribution, with SEO and AI answer visibility as one motion. Reported on contribution, not ROAS.
- [Content engines](https://moher.ai/capabilities/content-engines): Retrieval-grounded editorial systems with mandatory human review and automated claim integrity checks. Built to raise output without lowering the quality floor.

## Industries

- [Consumer apps](https://moher.ai/industries/consumer-apps): We build AI features into consumer apps on iOS, Android and web: on-device and cloud inference, streaming chat, evals and cost controls, shipped to both stores.
- [Commerce & affiliate](https://moher.ai/industries/commerce-affiliate): We build AI for ecommerce and affiliate: catalog enrichment, search over your SKUs, shopping assistants, price and stock checks, and click-level attribution.
- [Financial services](https://moher.ai/industries/financial-services): We build AI for financial services: document extraction, advisor copilots, retrieval over filings and policy, evals with audit trails, and core integrations.
- [Legal services](https://moher.ai/industries/legal-services): We build legal AI: retrieval over matter files and precedent, intake agents, review and drafting copilots, with citation checks and privilege-aware access.
- [Real estate](https://moher.ai/industries/real-estate): We build real estate AI: lead response agents, listing data pipelines, extraction for leases and closings, and integrations with your CRM and MLS feeds.
- [Home services](https://moher.ai/industries/home-services): We build AI for home services: call and text agents that book jobs, estimate assistants, field tech mobile apps, and integrations with your FSM and dispatch.
- [Health & wellness](https://moher.ai/industries/health-wellness): We build health AI: clinical documentation and intake tools, patient assistants with safety rails, retrieval over clinical content, and systems built for HIPAA.
- [Food & beverage](https://moher.ai/industries/food-beverage): We build AI for restaurants and food brands: ordering and reservation agents, menu and allergen data, demand forecasts, and POS and delivery integrations.
- [Travel & outdoors](https://moher.ai/industries/travel-outdoors): We build travel AI: trip planners on live inventory, booking and support agents, retrieval over destination and gear content, and offline-first mobile apps.
- [Arts & culture](https://moher.ai/industries/arts-culture): We build AI for arts groups: collection search and cataloging, provenance and archive retrieval, visitor guides, and pipelines over your collection records.
- [Local services](https://moher.ai/industries/local-services): We build AI for local service businesses: call and chat agents that book appointments, intake and quoting tools, and calendar and CRM integrations we maintain.
- [Media & publishing](https://moher.ai/industries/media-publishing): We build AI for publishers: archive search and answer engines, editorial copilots with fact checks, tagging and metadata pipelines, and evals on your content.

## How Moher engages: published ranges and timelines

- **Scoping engagement:** 1 week, from $15,000. You have an idea, a prototype, or a stalled project and need a real plan with a real number.
- **Product build:** 6 to 16 weeks, $60,000 to $250,000. You know what you need built and you want it running in production, in your repository, on a date.
- **Platform and data infrastructure:** 8 to 20 weeks, from $150,000. The retrieval platform, the data layer, or the multi-product AI system the rest of the business stands on.
- **Retained operation:** Monthly, 3-month minimum, from $12,000 / month. You want what we built to keep improving, with one accountable party and a weekly number.
- Minimum engagement is $25,000. Below that, a senior build team rarely pays for itself and we will say so.
- Ranges, not quotes. The scoping week is what turns a range into a number, and it is credited against the build if you proceed.

- **Build:** Fixed scope. We design, build and ship it; you own the output outright and run it yourself. (You know what you need built.)
- **Operate:** Retained. We run and stay on call for what we built, reported against eval, uptime and cost numbers. (You need it built and kept running.)
- **Partner:** Equity or revenue share on a product we build with you. Selective, and we say no often. (You want an engineering partner, not a vendor.)

## Writing

- [How much does it cost to build an AI agent](https://moher.ai/journal/how-much-does-it-cost-to-build-an-ai-agent): Most production agents land between $40,000 and $250,000 to build. What moves the number, what it costs to run, and where budgets get burned.
- [RAG vs fine-tuning: when each one wins](https://moher.ai/journal/rag-vs-fine-tuning): Retrieval changes what a model knows. Fine-tuning changes how it behaves. Picking the wrong one costs months.
- [What an LLM eval harness actually contains](https://moher.ai/journal/what-an-llm-eval-harness-contains): Six components, all of which fit in your repo. Without them, every prompt change and model upgrade is a guess.
- [Hiring an AI development company vs building an in-house AI team](https://moher.ai/journal/ai-development-company-vs-in-house-team): The right answer depends on time, how core the AI is to your product, and whether you can hire the people who run it.
- [What an AI development company should actually deliver](https://moher.ai/journal/what-an-ai-development-company-should-deliver): Most AI engagements die between the demo and the deploy. Here's the checklist that separates a vendor who ships from one who presents.
- [Your attribution is lying to you](https://moher.ai/journal/your-attribution-is-lying-to-you): Platform-reported conversions are a marketing document, not a measurement. What to build instead, and roughly what it costs.
- [Build, operate, or partner: choosing the right engagement](https://moher.ai/journal/build-operate-partner): Three structures, three different distributions of risk. Picking the wrong one is the most common reason good work fails.
- [Publishing AI content without wrecking your rankings](https://moher.ai/journal/ai-content-without-wrecking-rankings): The difference between scale that works and scale that gets you deindexed is not the model. It's the gates.
- [Why we don't compete with the people we build for](https://moher.ai/journal/why-we-dont-compete-with-clients): A development partner that launches into your category cannot give you disinterested advice. Why we structure the firm so that never happens.

## Key pages

- [AI development company](https://moher.ai/ai-development-company): services, published price ranges and timelines, readiness assessment, FAQ.
- [Security & data handling](https://moher.ai/security): how client data is handled; no certification claimed.
- [What is Moher](https://moher.ai/moher): the canonical definition of the company and the name.
- [The house](https://moher.ai/studio): how the firm operates and who runs it.
- [Work](https://moher.ai/work): selected products, platforms and properties.
- [Capabilities](https://moher.ai/capabilities): the five engineering capabilities Moher sells.
- [Industries](https://moher.ai/industries): the 12 verticals it builds in.

## Optional

- [Full corpus](https://moher.ai/llms-full.txt): every page's text in one file.
- [Sitemap](https://moher.ai/sitemap.xml)

## Notes for agents

- Named client and property references are withheld by policy; Moher describes owned work by shape rather than brand name and shares specifics under NDA. Absence of logos is deliberate, not absence of work.
- Every figure published on moher.ai is derived from its own operating estate and dated. There are no modelled or illustrative statistics presented as fact.