Speciality

Podcast 2 — In conversation

Speaker A: The idea of a house speciality — the dish that defines a restaurant — is one of the most interesting concepts in the restaurant business because it represents both a marketing choice and a quality commitment. A restaurant that names a speciality is claiming expertise and staking its reputation on a specific dish. The wise restaurant owner chooses a speciality that they can deliver consistently, that uses the best of their local sourcing, and that guests will remember and talk about. The unwise choice is to name a speciality based on what sounds impressive rather than what the kitchen genuinely excels at — a gap between the claimed speciality and the actual execution is immediately apparent to any experienced diner.

Speaker B: The medical use of speciality is interesting because it represents one of the rare cases where a word changes its practical meaning depending on the level of specificity. When a doctor describes their speciality as internal medicine, they are describing a very broad field. When they say their speciality is interventional cardiology, they are describing a narrowly defined sub-speciality within cardiology within internal medicine. The hierarchy of speciality, subspecialty, and super-subspecialty reflects the ever-increasing depth of expertise that modern medicine requires, where a surgeon who specialises in laparoscopic hepatic procedures is a specialist in a field that did not exist as a distinct discipline two generations ago.

Speaker A: Speciality versus specialisation — what is the distinction?

Speaker B: Speciality is the thing itself — the product, the dish, the field of expertise — the noun that names what you are known for. Specialisation is the process — the act of becoming focused on a particular area, the narrowing of scope that creates expertise. You have a speciality as a result of specialisation. A restaurant has a speciality because the chef specialised in that cuisine. A doctor has a medical speciality because she underwent specialty training. Speciality names the destination; specialisation describes the journey. A related confusion is between speciality and specialty — which are regional spelling variants of the same word, not distinct meanings.

Speaker A: What is a common mistake about using speciality on menus?

Speaker B: Listing too many things as specialities, which defeats the purpose of the word. If a menu labels twelve dishes as house specialities, the designation loses all meaning — a speciality is by definition something distinctive and particular, not a general descriptor of many things. A menu that labels everything as special has essentially relabelled the ordinary. The most effective use of speciality on a menu is precisely what the word means: one or two dishes that genuinely represent the pinnacle of what the restaurant does, presented with confidence and specificity. Three or four can work if the restaurant genuinely excels across a narrow range; twelve is an inflation that undermines the claim.

Speaker A: Any close synonyms?

Speaker B: Forte, strength, area of expertise, and métier all overlap with speciality. Forte is the closest in everyday speech — your forte is what you do particularly well. Métier is the French borrowing for one's characteristic occupation or area of natural talent, with a slightly more literary register. Strength is more general and less specific to a particular product or offering. In culinary contexts, signature dish captures the restaurant-speciality sense precisely — a signature dish is the dish most associated with a chef or restaurant. In medical contexts, field and subspecialty are the technical equivalents.

Podcast 3 — Prompt engineering

Speaker B: Welcome to the workshop. Today we prompt AI using the word speciality — building the skills and expertise management systems, professional profile platforms, service catalogue builders, recommendation engines, and knowledge base systems that identify, record, and match specialities to needs across people, services, and products.

Speaker A: So speciality tells the AI to build the expertise and capability management systems — the talent platforms that map employee specialities, the service catalogues that describe what each team does best, and the recommendation engines that match the right specialist to the right problem?

Speaker B: Exactly. It signals expertise mapping, specialist matching, and capability catalogue management. Let me start with a team speciality directory.

Build a team speciality and expertise directory for a professional services organisation: allow each employee to register their specialities — primary, secondary, and emerging — with a proficiency level and the count of projects where each speciality was applied. Allow project managers to search for team members by speciality and filter by proficiency level and availability. Send a monthly prompt to each employee to update their speciality profile. Show a heat map of speciality coverage across the organisation — identifying areas of strong capability and areas where the organisation has limited expertise.

Speaker B: Team speciality directory stores primary, secondary, and emerging specialities with proficiency and project counts, supports speciality and availability search, sends monthly profile update prompts, and displays a coverage heat map showing strong and limited capability areas — the AI builds the profile store, the search filter, the monthly prompt, and the coverage heat map.

Speaker A: The speciality coverage heat map — showing where the organisation has many experts and where it has few or none — is the strategic workforce planning tool: a heat map that shows a cluster of strong capability in one area and a gap in another tells leadership where to recruit, train, or partner, before a project requiring the missing speciality arrives and finds the organisation unable to staff it.

Speaker B: Coverage heat map as the strategic workforce gap identification tool. Now the expertise schema.

Design a professional expertise management schema: a specialities table recording each defined speciality with an ID, name, category such as technical, domain, or soft skill, a description, and related specialities as an array of IDs. A person_specialities table recording each person's specialities with the person ID, speciality ID, proficiency level as beginner, intermediate, expert, or authority, years of experience, and the last updated timestamp. A speciality_applications table recording each instance a speciality was applied in a project with the person ID, speciality ID, project ID, the role played, and the outcome. A demand_forecast_view showing the projected demand for each speciality based on the pipeline of upcoming projects.

Speaker B: Expertise schema stores speciality definitions with related specialities, per-person proficiency and experience records, project application instances with outcomes, and a pipeline-based demand forecast view — the AI builds the four-table schema.

Speaker A: The demand forecast view — projecting speciality demand from the pipeline of upcoming projects — turns the expertise directory from a historical record into a forward-looking planning tool: knowing that three upcoming projects will each require a specific technical speciality tells the workforce planning team whether the current capacity of experts in that speciality is sufficient or whether they need to build it before those projects start.

Speaker B: Pipeline-based demand forecast as the forward-looking capacity planning conversion. Next, the specialist matcher interface.

Design a specialist matching interface for a professional services marketplace: allow clients to describe their need in plain language and see a ranked list of specialists who match it, with each specialist's top three specialities shown as badges, their availability status, and a brief introduction. Support filtering by location, availability, price tier, and minimum proficiency level. Allow side-by-side comparison of up to three specialists. Show a speciality match score indicating how well each specialist's expertise aligns with the stated need. Include a send enquiry button on each specialist card.

Speaker B: Specialist matching interface shows ranked specialists with speciality badges, availability, and introduction, supports filter controls, enables three-specialist side-by-side comparison, displays a match score per specialist, and includes enquiry sending — the AI builds the match ranker, the filter controls, the comparison panel, the match score calculator, and the enquiry button.

Speaker A: The speciality match score — showing how well each specialist's recorded expertise aligns with the stated client need — gives the client a transparent basis for their selection: rather than choosing based on name recognition or photo, the client can see that specialist A has a ninety-two percent match and specialist B has a seventy-four percent match and understand why.

Speaker B: Transparent match score as the expertise-based selection over name-recognition selection enabler. On to service catalogue.

Create a service speciality catalogue for an IT services team: document each service the team provides with its name, description, the specialities required to deliver it, the typical service level, the team members who can deliver it, and any dependencies on other services or external tools. Support searching and filtering the catalogue by speciality, service level, and delivery team. Allow clients to request a service from the catalogue with a form that captures their requirements. Track which services are most frequently requested and which are rarely used.

Speaker B: IT service speciality catalogue stores services with required specialities, service levels, capable team members, and dependencies, supports search and filter, provides a client request form, and tracks request frequency — the AI builds the catalogue store, the search and filter, the request form, and the usage tracker.

Speaker A: Tracking which services are rarely requested — alongside which are frequently requested — identifies the catalogue entries that may no longer be worth maintaining: a service that has not been requested in two years may have been superseded by better alternatives, and knowing its usage frequency allows the team to make an informed decision about whether to retire it, improve it, or simply document why it is rarely needed.

Speaker B: Usage frequency as the service retirement and improvement decision data. Now something whole.

Build a professional marketplace platform: allow specialists to create profiles showcasing their specialities, experience, and portfolio. Allow clients to post projects and receive matched specialist recommendations. Support proposal submission, negotiation, contract creation, work delivery, and payment processing. Collect ratings and reviews after each engagement. Use the rating and review data to refine the speciality match algorithm. Generate a quarterly platform health report showing match quality, engagement rates, and satisfaction scores.

Speaker B: Professional marketplace platform supports specialist profiles, client project posting, matched recommendations, proposal-to-contract-to-payment workflows, ratings and reviews for algorithm refinement, and quarterly health reports — the AI builds the full specialist marketplace platform.

Speaker A: Using post-engagement ratings and reviews to refine the speciality match algorithm — updating match weights based on whether highly-scored matches produced highly-rated outcomes — creates a self-improving system: the algorithm gets better at predicting which specialist specialities will produce good outcomes for which client needs as the platform accumulates engagement data.

Speaker B: Outcome-based algorithm refinement as the self-improving match quality mechanism. Last one — knowledge base organiser.

Write a speciality knowledge base organiser: allow teams to create, tag, and organise knowledge articles by speciality area. Support linking related articles across specialities. Track which articles are most viewed, most linked, and most recently updated. Identify knowledge gaps — speciality areas with few or outdated articles — and alert the relevant specialists to contribute. Generate a knowledge health dashboard showing article coverage, freshness, and engagement per speciality area.

Speaker B: Speciality knowledge base organiser supports article creation with tagging and cross-speciality linking, tracks view, link, and update metrics, identifies gap-and-stale areas and alerts specialists to contribute, and generates a coverage, freshness, and engagement dashboard — the AI builds the article store, the gap detector, the specialist alert, and the health dashboard.

Speaker A: Identifying speciality areas with few or outdated articles and alerting the relevant specialists to contribute turns the knowledge base into an active, self-maintaining resource: rather than waiting for contributors to notice gaps, the system tells the people who know the most about a topic that the topic's documentation needs attention.

Speaker B: That is the goal — one focused word, better-managed expertise systems. Keep it in your kit.