EhlyTECH — Cross-Language AI Service Orchestration

The description states that EhlyTECH is an AI-assisted local service orchestration platform. It supports multiple input methods for user needs: search, category selection, natural-language descriptions, and image-based signals.

It claims to manage the full lifecycle of a local service request:

• Customer need
• Service routing
• Workflow planning (planned or urgent)
• Dynamic form creation
• Expert job card generation
• Offer submission and translation
• Messaging between customer and expert
• Agreement, reminders, completion, and rating

The platform also supports users acting as both customers and experts.

It uses a staged decision architecture:

1. Dictionary matching
2. Vector / semantic matching
3. Intent routing
4. Controlled LLM fallback
5. FAQ/chat/unsupported-service handling

Not evidenced: actual product functionality beyond the Build Week demo, or whether this is a live product or prototype.

The author states that EhlyTECH emerged from field experience as both a customer and service provider, identifying a core problem in local services — users do not think in catalog names but describe needs in natural language.

It positions itself as a solution to messy local service workflows by turning those descriptions into structured, validated workflows.

The platform claims to:

• Bridge customers and experts across languages
• Use AI to interpret needs and generate offers
• Support both text and image-based inputs
• Manage full service lifecycle from request to rating

It also references a long-term vision called HaYaT — an AI-assisted interface for managing daily life needs, including food ordering and supply delivery.

Inferred: the positioning reflects a move toward broader consumer-local service ecosystems. However, no evidence of market traction or customer feedback is provided.

The description states that EhlyTECH targets users who need local services — both customers seeking help and experts providing it.

It supports a dual role for users: they can be either customer or expert, allowing the platform to support both sides of the local service economy.

It also mentions:

• Founding Expert applications have started coming in through a website
• Applications are from different cities and service categories

Not evidenced: actual customer base, demographics, or segmentation data. The description does not clarify if this is B2C, B2B, or both.

The description states that EhlyTECH includes:

• Smart pricing infrastructure
• Geography, distance, city multiplier, category, and currency logic
• Expert statistics
• Founding Expert application and admin tracking infrastructure

It also mentions:

• Expert offer submission
• AI-assisted expert offer writing
• Offer translation into the receiver’s language

Not evidenced: actual pricing models, revenue streams, or monetization strategy. No information on whether users pay for services, experts are paid, or if there is a marketplace commission.

The existing product foundation uses:

• Flutter, Firebase, Firestore
• Firebase Cloud Functions
• Python backend logic
• Multilingual dictionaries
• Vector search
• Localization infrastructure
• Gemini-assisted production flows

For Build Week:

• GPT-5.6 was used in two roles:
  • Need Interpreter (structures natural-language needs)
  • Offer Bridge (translates expert offers into customer language while preserving facts like time, price, currency)

The architecture separates language intelligence from marketplace authority:

• GPT-5.6 handles interpretation and communication
• Deterministic layer controls service validation, workflow selection, and fact/claim validation

Tests and evidence include:

• 9 automated guardrail/unit/contract tests
• 2 verified live GPT-5.6 calls
• Live traces for Need Interpreter and Offer Bridge
• Recorded errors and fallback behavior

Inferred: the architecture is designed to reduce risk by not giving AI full control over marketplace decisions.

The description states:

• A pre-release website was launched to build expert supply side
• Founding Expert applications have started coming in from different cities and service categories
• The product already existed before Build Week as a Flutter/Firebase-based platform with core capabilities

Not evidenced: actual user base, revenue, or adoption metrics. No mention of active customers, paid users, or live marketplace activity beyond the website and expert applications.

The description does not provide any information about competitors or market positioning relative to other local service platforms or AI-assisted marketplaces.

Not evidenced: competitive landscape, differentiation, or market share claims.

• Operational Risk: The platform relies heavily on deterministic validation to control marketplace logic, which may limit scalability if the system becomes too rigid.
• AI Dependency Risk: While GPT-5.6 is used in limited roles, there’s no indication of how well it performs at scale or how it integrates with real-time data.
• Traction Gap: No evidence of revenue, customers, or live usage beyond the Build Week demo and expert applications.
• Market Maturity Risk: The long-term vision (HaYaT) is ambitious but unproven; no evidence of progress toward that goal.

Inferred: The platform may be in early development with limited commercial traction. The separation between AI and deterministic logic suggests a cautious approach, but also raises questions about how effectively it can scale or adapt to new use cases.

1. What is the current stage of the product beyond Build Week? Is there a live version or prototype?
2. How many actual expert applications have been received and how many are currently active?
3. What are the key performance indicators (KPIs) for service completion, customer satisfaction, and expert retention?
4. How does the deterministic layer handle edge cases or ambiguous inputs that don’t fit existing workflows?
5. Are there any plans to integrate with third-party payment systems or logistics providers?
6. What is the current cost structure of running the AI components versus the deterministic backend?
7. How do you plan to scale beyond Türkiye into other markets?