About Tiralis

Tiralis builds AI-powered compliance software for regulated industries, with an initial focus on life sciences.

Companies in these industries operate through controlled procedures, standards and records. Understanding how those sources relate, what changed and where a gap may exist still takes a great deal of manual work.

Tiralis uses language models, structured data and knowledge graphs to turn those sources into traceable information. We connect procedures, requirements, evidence and operating knowledge so that quality teams can use them in their daily work, with a clear path back to the source.

The work combines document processing, context engineering, model training and evaluation, structured extraction, human review and the product workflows that bring those parts together.

Tiralis is not a multi-tenant SaaS play. Each client receives a separate tenant deployment. The platform must work in the cloud, on premises and, where required, inside isolated environments, based on the client compliance requirements, and preferences.

The role

This is the first dedicated technical architect seat at Tiralis, and you will join as a member of Tiralis’s founding technical staff.

Architect here means you design the approach and build it. You would write production code across the stack while guiding application development day to day. You would turn broad direction from the CTO into technical plans, working software and a development approach that a small team can sustain.

The CTO owns the wider technical direction, client solution design, deployment strategy, support and people management. You would add the day-to-day focus needed to drive development while working closely with the CTO on the choices that shape the platform.

You would also work closely with the Product and ML teams. The Product team brings domain knowledge and subject-matter expertise. The ML team develops the context-engineering, model-training, evaluation and extraction methods behind the product. Your role is to help bring those parts together as dependable software.

The product is still early, and we are building toward MVP. Some significant choices remain open. Others will change as we test the product with users and learn what clients need. The outcome you own is broad: help build a sound, adaptable technical foundation and turn Tiralis’s AI, data and domain capabilities into a product that real quality teams can rely on.

Who you would be working with A small founding team with a bench of specialists. Product, domain, engineering, machine learning and commercial experience all have to be in the room for this to work.

- Domain and Product. A Product team that brings deep knowledge of quality engineering, auditing and regulated manufacturing. This includes a founder with fifteen years in quality engineering and auditing inside regulated manufacturing, formally certified in both, across production sites in Europe, North America and India. The side of the desk that gets inspected. The team also works with subject-matter experts in quality and regulation.
- Engineering. A hands-on technical founder with extensive startup experience across B2B and compliance-heavy domains, who owns development, delivery, client technical solutioning, and all things technical in-between.



You would work directly with them and provide the day-to-day technical focus for application development.
- Machine learning. A Senior Machine Learning Architect and ML team working on document processing, context engineering, model training and evaluation, structured extraction and knowledge pipelines. You would work closely with them to turn those capabilities into dependable product workflows.
- Commercial. A founder trained in computational biology who has founded and run companies in drug discovery and biotechnology, and who leads partner relationships. Real manufacturers and real documents come through this side.
- Specialists. Senior quality and regulatory professionals who review our data model and test the product, and machine-learning engineers working on large-scale extraction in other regulated industries.

The platform The platform is being built around:

- React and TypeScript
- Go services backed by PostgreSQL and Redis.
- REST, gRPC and WebSocket interfaces.
- AI and LLM pipelines for processing documents and producing structured compliance data.
- Context-engineering, model-training and evaluation workflows developed with the ML team.
- Knowledge-graph capabilities that connect source material, extracted knowledge and external requirements.
- Container-based deployment, with AWS as the current development environment.
- Terraform-managed AWS infrastructure.

We are designing the core data and knowledge layer to remain stable and reusable while application and UI modules develop around product and client needs. Over time, we expect to separate current operational state from historical data. Lakehouse and versioned-data patterns will help us preserve past versions of records and knowledge graphs for auditability and time-based views. Apache Iceberg is one option on that roadmap.

What you will work on

Your focus will change as the product develops, but the broad areas include:

- Guiding application design and daily engineering work.
- Building production software across frontend, backend and data.
- Working with the Product team and subject-matter experts to turn domain needs into product and technical choices.
- Working alongside the ML team on context engineering, model training, evaluation and pipeline development.
- Turning ML and LLM pipeline outputs into dependable product workflows.
- Designing the application, data and deployment services needed to run AI capabilities within client-controlled environments.
- Connecting documents, model outputs, structured records and knowledge graphs while preserving provenance and audit history.
- Keeping core data and knowledge interfaces stable and reusable.
- Supporting modular application and UI capabilities without causing drift in the core platform.
- Keeping the system portable across cloud, on-premises and isolated deployments.
- Improving security, access control, testing, releases and maintenance.
- Supporting early deployments and other technical work where the team needs help.

What we are looking for





- Experience designing and building production systems while remaining hands-on in the code.
- Strong full-stack engineering skills across web applications, services and data.
- The ability to guide technical planning, implementation and code review.
- Strong experience with lakehouse or versioned-data architecture, including historical state, schema change and audit needs.
- Experience designing and building containerised systems.
- A sound understanding of how code, dependencies, storage and infrastructure choices affect portability.
- Experience designing stable APIs, query interfaces and modular systems.
- The ability to work with rough requirements and make progress while important details remain open.
- Clear communication with engineers, product teams, founders and domain specialists.
- A record of learning unfamiliar systems and domains quickly.

Production experience with Go and React/TypeScript is preferred. We will also consider strong candidates with experience in comparable technologies who can learn the current stack quickly. We care more about what you have built, the choices you owned and how you work than degrees, titles or a fixed number of years.

Nice to have

- Regulated, audited or compliance-led products, preferably in life sciences.
- Experience building applications around ML or LLM pipelines, including context engineering, model training, evaluation or human review.
- Knowledge graphs, ontologies or linked-data systems.
- Apache Iceberg or another open table format.
- AWS and Terraform.
- Systems deployed on premises or in isolated environments.
- Experience building a product from an early stage.

AI-assisted development We expect engineers to use modern AI coding tools such as Claude Code or Codex. Experience using them effectively is preferred and will form part of the interview process.

What you should know going in

- Tiralis is early and still building toward MVP.
- Product priorities and technical choices will develop as we learn.
- AI and LLM capabilities form a core part of the product.
- Model execution, context data and pipeline outputs may need to remain within the client’s deployment boundary.
- The technical, Product and ML teams will develop the system together, with close links between model behaviour, data design and user experience.
- Each billing client receives a separate tenant deployment, either on prem, cloud or a hybrid deployment.
- AWS supports our current development environment, but the platform must remain infrastructure agnostic.
- Customer environments may restrict public cloud services and internet access.

How we hire The process begins with an introductory call, followed by a practical exercise based on a real Tiralis problem. The exercise is scoped to four or five focused hours. If you reach five hours, stop and tell us where you stopped. We will then spend about two hours reviewing your work, the choices you made and the options you rejected. This discussion carries more weight than polish.

AI coding tools are allowed and expected. Tell us what you used and how it affected your work.

One-to-one conversations with founders may follow the technical review. The process concludes with an HR discussion covering fit and practical terms.

Practicalities

Location: Remote within India

Type: Employee

Compensation: Negotiable; Cash and equity, with vesting

Reports to: CTO

Start: ASAP

📌 Senior Member of Technical Staff (India)
🏢 Tiralis AI
📍 India