Meet Tanu: The AI Interviewer Rethinking How We Evaluate Engineering Talent

Meet Tanu: The AI Interviewer Rethinking How We Evaluate Engineering Talent

We have all heard stories about talented engineers who struggle in interviews, and candidates who perform well in coding tests but find real world engineering much harder.

It got us thinking. Are technical interviews really measuring engineering ability, or are they measuring how well someone has learned to interview?

At Mindfries, this question is at the heart of what we are building. We want to make technical interviews more meaningful, more practical, and closer to the work engineers actually do.

That is why we built Tanu.

So, who is Tanu?

Tanu is the AI interviewer at Mindfries.

Think of Tanu as a way to explore how someone approaches a problem, explains their thinking, investigates an issue, and makes technical decisions. The goal is to look beyond whether a candidate can produce the right answer and understand more about how they get there.

Because being a good engineer is rarely just about writing code.

Sometimes it means finding a bug that nobody else can explain. Sometimes it means making a difficult architectural decision. Sometimes it means working with an unfamiliar system, understanding what is going wrong, and figuring out what to do next.

These are the kinds of skills we believe technical interviews should pay more attention to.

Engineering has changed. Interviews should change too.

Think about the range of engineering roles companies hire for today.

A Forward Deployed Engineer might need to understand a customer's problem, work with unfamiliar systems, and build a solution under real constraints.

A security researcher might need to investigate unexpected behaviour and understand where a system could be vulnerable.

An AI engineer might need to understand model behaviour, evaluate outputs, and make decisions about reliability, latency, and cost.

A Site Reliability Engineer might need to investigate an incident and work out how to prevent it from happening again.

These roles have very different responsibilities. Yet hiring processes can still rely heavily on the same familiar coding questions.

Coding assessments have their place. They can be useful for measuring specific programming skills. But they do not always tell the whole story.

We believe the interview should reflect the role, not force every role into the same format.

One interviewer, many kinds of engineering challenges

Our vision for Tanu is to make technical assessment more adaptable to the skills each role actually requires.

Here are ten areas where we believe a more practical approach to interviewing can make a difference.

1. Forward Deployed Engineers

These engineers work close to customers and solve problems in real environments. They need technical depth, curiosity, and the ability to turn unclear requirements into working solutions.

2. Reverse Engineers

Reverse engineers investigate how software and systems behave. Their work depends on careful observation, forming hypotheses, and reasoning from evidence.

3. AI Research Engineers

These engineers experiment with new approaches, study results, and turn research ideas into working implementations. Their work requires both technical understanding and a willingness to investigate what does not work.

4. Site Reliability Engineers

Reliability engineers help keep systems available and dependable. Incident investigation, debugging, monitoring, and thoughtful trade offs are all part of the job.

5. Security Researchers

Security researchers need to think critically about how systems can fail or be misused. Their assessments should explore investigation, threat awareness, and responsible security reasoning.

6. AI and LLM Engineers

Building useful AI products involves much more than connecting an application to a model. Engineers need to consider output quality, evaluation, retrieval, reliability, latency, and cost.

7. Developer Infrastructure Engineers

These engineers create the tools and platforms that help other developers do their jobs. Their work can involve build systems, automation, internal tooling, and improving the developer experience.

8. Systems Engineers

Systems engineers need to understand how different components interact. They often work through performance issues, architecture decisions, resource constraints, and unexpected behaviour.

9. Data and ML Infrastructure Engineers

These engineers build and maintain the systems behind data pipelines and machine learning workflows. Reliability, scalability, data quality, and resource management are central to their work.

10. Solutions Engineers

Solutions engineers connect technical products with real customer needs. They need to understand a problem, design an appropriate solution, explain their decisions, and work through practical constraints.

Each of these roles calls for a different mix of skills. Our goal is to help make those differences easier to explore through assessments designed around relevant engineering work.

We want interviews to feel more like engineering

Imagine being asked to investigate a bug instead of simply recalling a programming concept. Imagine discussing why one system design might work better than another, or explaining how you would approach an unfamiliar technical problem.

These conversations can reveal things that a single coding question might miss.

That is the direction we want to take with Tanu.

We want assessments to encourage candidates to think aloud, explain their reasoning, and demonstrate how they approach realistic challenges. We also want hiring teams to have a clearer structure for understanding what a candidate has demonstrated.

AI can help make these experiences more flexible and consistent, but it should not replace thoughtful human judgment. The quality of an assessment still depends on asking relevant questions, using fair criteria, and reviewing results in context.

What we are learning as we build Mindfries

Mindfries started with an idea that felt simple to us: technical hiring should do a better job of recognising real engineering ability.

We built our first version, put it in front of early users, and began learning from the experience. Building V1 gave us something concrete to test, question, and improve.

Now we are working towards V2.

For us, this is not simply about adding more features. It is about making Tanu more useful, expanding the kinds of engineering challenges we can support, and creating an assessment experience that better reflects what companies need from their engineers.

We know there is plenty to learn and improve. That is part of building something new. Every iteration gives us another opportunity to understand where technical interviews fall short and how we might make them better.

The future we believe in

We do not believe there is one perfect interview for every engineering role. We also do not believe an AI interviewer should claim to understand everything without limits.

What we do believe is that AI can help us ask better questions, explore a wider range of skills, and make technical assessments more relevant to the work people will actually do.

That is what we want Mindfries to stand for.

Tanu is more than a name on an interview screen. It represents our effort to rethink what technical hiring can look like when we focus on the person behind the code.

We are building Mindfries one step at a time, learning as we go, and working towards a future where engineering talent can be evaluated through what people can actually do.

Because the best interview is not necessarily the one with the hardest coding question. It is the one that helps you understand how someone thinks, builds, and solves problems.

If that sounds like a better way to hire engineers, we would love for you to follow our journey.

Visit mindfries.com to learn more about what we are building.