Interview Scorecards

How to Read Product Portfolios When Hiring UI/UX Designers From Bangalore's Startup Belt

A Bangalore employer framework for reading UI/UX portfolios as evidence of product judgment, research, systems thinking, collaboration, and shipped outcomes.

Sonal 16 August 2026 12 min read
Choose your PlaceMeRight path Candidates can open their dashboard. Employers can request a tech shortlist.

A polished UI/UX portfolio can be useful and still conceal the decision an employer needs to make.

The screens may look refined, the story may sound fluent, and the case study may show a complete process. None of those details automatically establishes what the designer personally discovered, decided, changed, shipped, or learned.

For Bangalore startups, product companies, and enterprise teams, portfolio review works best as evidence reconstruction.

Ground the search in a bounded Bangalore sample

On 28 July 2026, PlaceMeRight reviewed the newest 120 records returned by its Bangalore live-jobs endpoint. The records were published from 21 to 23 July, and the endpoint reported 1,733 Bangalore records at collection time.

Use PlaceMeRight's Bangalore technology recruitment context when turning this market observation into a role-specific designer search.

A conservative title-only design filter identified eight records:

  • Associate UX Designer;
  • Product Designer;
  • Lead Product Designer;
  • Product Designer;
  • UX Designer 4;
  • Senior Product Designer;
  • Senior Product Designer for Bangalore or Noida;
  • Senior Product Designer.

The employers included PlaySimple Games, Zeta Global, Uniqode, ShopDeck, Nutanix, Newton School, Level AI, and GreyLabs AI.

This is not a measure of designer supply or city demand. It is a newest-120 title sample, and it omits design work hidden inside other titles. Compensation, workplace type, and experience fields were unpopulated. Location was only Bangalore, so the data does not assign any role to Koramangala, HSR Layout, Indiranagar, Domlur, Bellandur, or another startup area.

Its value is narrower: employers are using UX, product, lead, and senior labels for different design mandates. The portfolio review must discover the mandate underneath the label.

Read the portfolio through an evidence ladder

Use six levels. A beautiful artefact can sit at a low level if the candidate cannot connect it to decisions.

Level 1: visual output

What was made?

Look for hierarchy, typography, spacing, interaction states, responsive behaviour, accessibility, motion, data display, and component consistency.

This level tests craft. It does not prove that the candidate solved the right problem.

Level 2: interaction logic

How does the user move through the product?

Ask about entry points, decision paths, empty states, errors, permissions, interruptions, undo, progressive disclosure, and recovery.

A designer who shows only the happy path has not yet demonstrated a complete interaction model.

Level 3: problem framing

Why did this work exist?

The candidate should distinguish a business request from a user problem, identify assumptions, and explain what was deliberately excluded.

Listen for a problem statement that changed as evidence emerged. A fixed statement followed by decorative research may indicate a retrospective process story.

Level 4: evidence and trade-offs

What changed the design?

Evidence may include interviews, support tickets, behavioural data, usability sessions, sales feedback, operational observation, domain constraints, or technical discovery.

The candidate should explain the quality and limits of the evidence. "Users wanted simplicity" is not enough.

Level 5: delivery and collaboration

What reached users, and how?

Look for engineering collaboration, component reuse, design QA, instrumentation, staged release, documentation, and the handling of constraints.

This level separates a concept portfolio from production ownership.

Level 6: outcome and learning

What happened after release?

The answer may be a measurable product result, a usability improvement, lower support burden, faster task completion, fewer errors, stronger adoption, or a decision to reverse course.

Not every employer can share a metric. A strong candidate can still describe the observation method, caveats, and next decision without inventing impact.

Reconstruct authorship before rating quality

Case studies often use "we". Ask the candidate to map the team:

| Role | Contribution to establish | | --- | --- | | candidate | research, interaction, visual, systems, facilitation, QA, or strategy | | design peers | critique, parallel flows, system ownership, research support | | product | prioritisation, problem definition, roadmap, measurement | | engineering | feasibility, architecture, component implementation, instrumentation | | data or research | evidence collection and interpretation | | leadership or client | constraints, approvals, commercial decisions |

Then ask:

  • Which decision was yours?
  • Which decision did you influence but not own?
  • Which recommendation was rejected?
  • Which artefact did another designer create?
  • What would your collaborators say you contributed?

Authorship is not about claiming credit. It is about predicting how the person will operate in the new team.

Compare current role signals

Current official job descriptions reveal different portfolio expectations.

DevRev's Bangalore Product Designer role emphasises complex product areas, AI-driven interaction, research, system-level problems, autonomy, and explicit trade-offs.

Glean's Bangalore role asks for an end-to-end product lifecycle, research, prototyping, detailed flows, design systems, roadmap partnership, B2B experience, and the ability to reason through technical concepts.

Instead's Bengaluru position combines complex tax workflows, rapid prototyping, a design system, engineering feasibility, customer support input, and collaboration with a distributed design team.

Sarvam's role asks for end-to-end product work, visual craft, motion, interaction, product judgment, and a portfolio showing real depth. Tellius asks a senior designer to handle AI, analytics, data-heavy interfaces, provenance, confidence, and non-deterministic states.

These roles do not define a single Bangalore standard. They illustrate five different portfolio questions:

1. Can the designer simplify a complex B2B workflow? 2. Can the designer shape an AI interaction whose output varies? 3. Can the designer maintain system coherence while shipping quickly? 4. Can the designer move between product judgment and visual craft? 5. Can the designer collaborate through implementation and learning?

Choose the questions that match the vacancy.

Read different product portfolios differently

Consumer portfolio

Look for motivation, comprehension, trust, habit, accessibility, and behaviour across repeated use.

Ask how the designer avoided optimising one conversion step at the expense of customer understanding or long-term value.

For gaming, commerce, education, or consumer finance, verify how experimentation and responsible design interacted.

B2B SaaS portfolio

Look for roles, permissions, configuration, dense information, onboarding, discoverability, and workflows that span several sessions or team members.

Ask how novice and expert users were served without creating two unrelated products.

Portfolio polish matters, but information architecture and operational edge cases often predict the actual work.

AI product portfolio

Static screens do not show enough.

Ask about:

  • uncertain or variable outputs;
  • user control and correction;
  • provenance and confidence;
  • latency and partial results;
  • prompt or instruction design;
  • unsafe or low-quality output;
  • fallback to deterministic workflows;
  • evaluation and feedback.

A candidate does not need to use every AI-design pattern. They should recognise that non-deterministic behaviour changes interaction design.

Data and analytics portfolio

Inspect hierarchy, comparison, time, filter state, units, missing data, confidence, annotation, export, and decision context.

Ask what the user does after viewing the dashboard. A portfolio that ends at visualisation may not demonstrate product consequence.

Regulated or high-trust portfolio

For finance, health, security, or identity work, look for consent, auditability, explanation, privacy, permissions, error recovery, and escalation.

Do not reward a candidate for disclosing confidential material. Strong portfolios can mask sensitive data and still explain the design reasoning.

Run a 45-minute portfolio evidence session

Minutes 0-5: choose one case

Ask the candidate to select the project most relevant to the role. Do not require a complete portfolio tour.

State that the panel will interrupt to understand decisions. This reduces pressure to perform a memorised presentation.

Minutes 5-15: frame the system

Ask for the user, business context, product maturity, team, candidate role, constraints, and starting evidence.

Capture unknowns rather than penalising them.

Minutes 15-27: trace one decision

Choose one important flow or interaction. Move from initial assumption to evidence, alternatives, decision, implementation, and result.

Use neutral probes:

  • What did you know at this point?
  • What alternative did you discard?
  • Who disagreed?
  • What technical constraint changed the design?
  • What state is missing from this screen?
  • What happened after release?

Minutes 27-37: change the constraint

Introduce a realistic change.

For example, the organisation cannot add another support agent, the AI result may be delayed, the user has a different permission, or the mobile connection drops halfway through.

Evaluate adaptation, not speed.

Minutes 37-45: reflect

Ask what the designer would change now, what evidence remains weak, and which skill they would need from the new team.

Self-critique should be specific. "I would conduct more research" is weaker than naming the unanswered decision and the method that could resolve it.

Do not overvalue the case-study template

Many portfolios use the same sequence: discover, define, ideate, prototype, test.

A real project may not unfold that neatly.

Good evidence can include:

  • revisiting the problem after engineering discovery;
  • shipping a limited version to learn;
  • using support evidence when direct research was unavailable;
  • changing a component system after repeated exceptions;
  • rejecting a requested feature;
  • discovering that the main issue was operational rather than interface-based.

Evaluate the integrity of the reasoning, not compliance with a fashionable process diagram.

Use a narrow, paid work sample when needed

A portfolio may leave a specific evidence gap. Use a small paid exercise only when it tests that gap.

For example:

> A B2B administrator needs to review AI-generated recommendations for several accounts. Some recommendations are incomplete, some conflict with policy, and some need another person's approval. Sketch the information and decision flow. State assumptions and open questions.

Limit the exercise to a few hours. Do not ask for a complete redesign of the employer's live product.

Evaluate:

  • quality of questions;
  • state and role modelling;
  • information hierarchy;
  • error and escalation paths;
  • testable assumptions;
  • communication of unfinished thinking.

The exercise should not be used as free product work.

Build a portfolio scorecard

| Dimension | Weight | Evidence | | --- | ---: | --- | | problem framing | 15 | case setup and reframing | | user and domain evidence | 15 | research quality and limits | | interaction design | 20 | flow, states, and recovery | | visual and systems craft | 15 | hierarchy, components, accessibility | | product judgment | 15 | trade-offs and prioritisation | | delivery partnership | 10 | engineering, QA, release | | authorship and communication | 10 | contribution map and reflection |

Define anchors before the panel meets.

A high score in product judgment may mean the candidate explains what not to build, connects the decision to evidence, and names the cost of the chosen path. A mid score may describe alternatives but not their consequences. A low score may present preference as user need.

Every score should quote an observed example.

Account for Bangalore workplace reality

Bangalore's startup ecosystem spans HSR Layout, Koramangala, Indiranagar, Domlur, Bellandur, Whitefield, and other corridors, but a candidate's portfolio does not reveal commute feasibility.

The role brief should provide:

  • exact office location;
  • expected office days;
  • whether design critiques happen in person;
  • where product and engineering partners sit;
  • customer or research time-zone requirements;
  • whether the designer will be solo, embedded, or part of a design team.

Do not describe a role as part of the "startup belt" unless the office and operating context support that description.

Official Karnataka sources document a substantial startup and innovation ecosystem. The 2026 Startup India state report describes KITS as the nodal agency and notes the state's updated Startup Policy 2025-30. That is ecosystem context, not evidence that every Bangalore designer has startup experience.

Avoid common portfolio-reading errors

  • Equating visual polish with product depth. Score craft and reasoning separately.
  • Rewarding a confident narrative. Interrupt the sequence and inspect evidence.
  • Penalising protected information. Ask for decision logic, not confidential data.
  • Assuming title consistency. UX, product, lead, and senior labels vary by employer.
  • Ignoring implementation. Ask what shipped and how quality was checked.
  • Using one standard for every product. Consumer, B2B, AI, data, and regulated work differ.
  • Treating Figma speed as seniority. Seniority includes judgment, influence, and systems thinking.

Quick checklist

  • Define the product and design boundary before sourcing.
  • Choose the portfolio evidence most relevant to the role.
  • Reconstruct personal authorship.
  • Trace one decision from evidence to release.
  • Inspect unhappy paths and edge states.
  • Test adaptation with one changed constraint.
  • Separate visual craft from product judgment.
  • Use a paid, bounded exercise only for a real evidence gap.
  • Score with behavioural anchors.
  • State the exact Bangalore workplace and team model.

Frequently asked questions

How many portfolio projects should a candidate present?

One deeply examined relevant case and a shorter contrasting example are often more useful than a full tour. Depth matters more than slide count.

Should every UI/UX designer show research?

The candidate should show how user or domain evidence informed decisions. The method can vary by role and access. Evaluate evidence quality and honesty, not a mandatory research ritual.

Can we hire a strong visual designer for a product-design role?

Yes, if the role's missing capabilities can be supported and the candidate demonstrates the required interaction, product, and collaboration potential. Do not silently assume visual craft covers those areas.

How should we assess confidential enterprise work?

Allow redacted or reconstructed material. Ask about constraints, alternatives, decisions, collaboration, and learning without requesting protected data or proprietary screens.

Does startup experience matter for a Bangalore startup?

It may help with ambiguity and speed, but evaluate the actual operating behaviours. Candidates from enterprise, agency, or services environments may have transferable evidence.

What is a useful senior-designer signal?

Look for consistent judgment across problem framing, systems, trade-offs, collaboration, and post-release learning, plus the ability to raise the quality of others.

Conclusion

A portfolio should not be scored as a gallery.

It is a record of claims that the panel can examine: the problem, evidence, decision, craft, collaboration, release, outcome, and candidate's own contribution.

Bangalore employers that use this evidence ladder can compare designers with more consistency while respecting the very different ways product work is documented across startups and established teams.

CTA

PlaceMeRight helps employers calibrate product-design roles, source relevant Bangalore profiles, structure portfolio review, and keep interview evidence aligned with the actual product mandate.

Share your UI/UX designer hiring requirement

References

Our Tech Recruitment Services by City

Looking to hire vetted software engineers or scale your engineering teams in India's leading tech hubs? Explore our localized recruitment services:

Sonal

PlaceMeRight recruitment insights

Practical hiring notes, scorecards, and recruitment operating guidance from the PlaceMeRight team.

Ready to Hire Developers?

Talk to PlaceMeRight for urgent developer shortlists and practical IT recruitment support.

Talk to Hiring Consultant