Shafeen Alam
All work
Interpres SecurityExplainable AI decision supportFlagship

Designing a governed, explainable AI assistant for threat & exposure management

Security analysts spent ~45 minutes per incident synthesizing exposure data across fragmented dashboards. I led the 01 design of Interpres's first AI assistant, which cut time-to-insight to 27 minutes and raised analyst trust from 62% to 81%.

Role
Design Lead, AI/UX
Timeline
2 weeks, 01 (2024)
The company was later acquired.
Team
PM, Eng, CTO
Scope
Strategy, conversational AI, data viz, design system
case.md
$ cat outcome
time_to_insight: 45 → 27 min
analyst_trust: 62% → 81%
Interpres Assistant chat interface with a structured response card
Assistant chat with a structured response card
Why it mattered

Every minute spent stitching dashboards together was a minute an exposure stayed unaddressed.

60-second read

Problem

Analysts spent about 45 minutes per incident synthesizing exposure data across fragmented dashboards.

System decision

Replace open-ended prose with a schema-constrained response structure that is evidence-linked and confidence-scored.

My contribution

Led the 0 to 1 design end to end: interaction model, evidence layer, safety and uncertainty states, design system.

Impact

Time to first insight 45 to 27 minutes. Analyst trust 62% to 81%.

Applied AI at a glance

An assistant that finds context, answers follow-ups, and acts only with approval.

Search

Retrieves the product and security context.

  • ContextInherits the selected exposure, asset, or technique.
  • Search and synthesisRetrieves and synthesizes exposure, control, and threat-intelligence data.

Chat

Lets analysts question and clarify that context.

  • AI chatFollow-up questions keep the investigation context.
  • TrustResponses expose evidence, provenance, confidence, freshness, and limitations.
  • RecoveryLow-confidence, slow, or incomplete responses offer a useful way forward.

Agent

Proposes or performs bounded, traceable actions, with human approval.

  • Tool useFilters, searches, and maps exposures, with every action traceable.
  • Human controlConsequential changes require explicit confirmation.
EvaluationTrust, time to insight, evidence validation, uncertainty, acceptance, and overrides.
  • 45 27 min to first insight
  • 62 81% analyst trust
  • 78% evidence validation
  • 8% AI uncertain rate
Shipped in BetaIllustrative recreationConcept extensionHow this page labels each piece of work. Everything above shipped or entered Beta. Metrics are from Beta analytics.

01 · Shipped

70%
of analysts opened it daily
45 27 min
time to first insight
2
customer wins enabled during Beta

A working 01 prototype, designed and validated in two weeks, that moved into Beta with real analysts, not a concept deck. It became the entry point to their workday, and a factor in closing new business.

The Interpres assistant dock open beside the exposure workspace, the surface analysts opened at the start of every shift
The surface 70% of analysts opened daily

02 · The problem

Security teams work across fragmented tools: alerts in one, exposure graphs in another, control coverage elsewhere. Even where the underlying data was rich, synthesis was entirely manual. Across 12 analyst interviews and 3 SOC manager sessions, the gap wasn't intelligence. It was structured, contextual reasoning.

“I don't trust automated suggestions unless I see the source.”
“Switching tools breaks my thinking.”

03 · What failed

First pass

Open-ended prose responses. In a security context that was unacceptable: hallucination risk, slow to scan under time pressure, and analysts finished reading unsure what to actually do next.

Illustrative, not the shipped copy

“Based on the available data, this could indicate a potential control gap related to access management, though further investigation may be needed to confirm the specific technique and assess the overall exposure level across affected systems...”

What replaced it

A schema-constrained response structure, rendered as cards instead of prose, with a “Why this matters” block on every answer.

EXPOSURE CONTROL GAP MITRE TECHNIQUE RECOMMENDED ACTION

04 · Where it lives

The assistant docks into the existing investigation rather than becoming a place you go.

Exposure dashboards, asset details, and MITRE views stay visible while the assistant works with whatever is selected. The dock inherits the active exposure, asset, or technique, so a question never starts from nothing.

Evidence links return focus to the source view instead of opening a separate AI destination. That single decision is what keeps it an assistant rather than a second product.

The Interpres campaign view for Operation CuckooBees, showing the description, a relationship graph linking techniques, software, threat groups, vulnerabilities and detections, and a properties panel
The investigation the analyst is already in. Campaign, relationship graph, defense surface, and associated threat elements.↗ open full size

The sequence it supports

It starts in place. The selected surface supplies the starting context for the question.

01

Clarify the score

Follow-up questions keep the current object and explain product language.

An assistant response explaining how the Priority Score is calculated, breaking it into its Visibility and Detections inputs
Product language explained in place. The analyst asks how a score is calculated and gets its two inputs, rather than a definition page in another tab.↗ open full size
02

Inspect why it matters

Evidence and impact stay inspectable before the analyst chooses an action.

The Why This Matters panel showing the MITRE ATT&CK vector Access Token Manipulation T1134 with an 85 confidence ring, an explanation of the exposure, a Learn more link and an urgent action callout, beside a role selector offering Analyst, Responder and CISO framings
Evidence, impact, and confidence in one panel, plus the role selector. The same finding changes depth and language for an analyst, a responder, or a CISO.↗ open full size
03

Recover safely

Low confidence offers refinement or an alternate query rather than guessing.

The AI uncertain state warning that the answer is based on limited data, offering two recommended recovery paths: refine your question by narrowing scope, or try an alternative query with worked examples
The refusal, with a way forward. Two recovery paths, each with a worked example, instead of a guessed answer the analyst might act on.↗ open full size

The patterns in place

Confirmation dialog warning that an action will modify firewall policy, with Confirm and Cancel
Human-in-the-loop. Any action touching a live control stops for explicit confirmation before it executes.
AI Uncertain state offering recovery paths: refine your question, or try an alternative query
The “AI Uncertain” pattern. Low confidence surfaces recovery paths instead of a guessed answer.
Still working on it fallback state with options to continue waiting or stop
Fallback for slow queries. The system stays legible instead of stalling silently.

05 · Live recreation

Not a screenshot. A working recreation of the reasoning structure and the evidence layer, with the two branches that matter most: uncertain and requires confirmation. Pick the prompt, take either follow-up, then try the sensitive action below.

Interpres Assistant · recreation Interactive
Try it · two branches to test
  • Uncertainty and recoveryTake the second follow-up. The assistant says what it cannot establish and offers a better question.
  • Consequential actionApply the recommended change in the panel below. It stops for one explicit confirmation.
Recreation · not live data

The second branch: a sensitive action

Every action that touches a live control stops here. One action, one confirmation, every time, and the dialog states the blast radius in the analyst's terms rather than the system's.

Try it. Confirm, or cancel and keep the existing rule.

Human-in-the-loop
Recommended action: enable attachment sandboxing for Finance OU.

One request, end to end

The patterns on this page, as a single request. Search brings the context, chat refines it, and the agent step stops for a person.

  1. 01

    Ask

    The analyst asks about this week’s phishing briefing.

    Shipped
  2. 02

    Inherit context

    The dock picks up the selected exposure, asset, or technique.

    Shipped
  3. 03

    Search

    It retrieves exposure, control, and threat-intelligence data.

    Shipped
  4. 04

    Show progress

    Status messages say what it is checking while it works.

    Shipped
  5. 05

    Recommend

    An answer arrives with linked evidence and a confidence score.

    Shipped
  6. 06

    Inspect

    The analyst checks sources, confidence, limitations, and scope.

    Shipped
  7. 07

    Confirm

    A consequential action stops for one explicit confirmation.

    A person decides hereShipped
  8. 08

    Result and recovery

    The result, a trace of what changed, and a way to recover.

    Concept extension

Steps 1 to 7 describe the shipped interaction model, and the recreation above lets you try most of them. Step 8 is a concept extension based on the shipped interaction model, and the recreation shows a simplified applied state. Autonomy stayed bounded: one action, one confirmation, every time.

06 · Confidence people
can use

A percentage helps only when someone knows what it means, can verify its basis, and understands the next action.

Precision without meaning creates false assurance, which in a security tool is worse than no number at all. So I treated the confidence score as three separate design problems rather than one label.

1

Score definition

Name what it represents

A number is ambiguous until you say which thing it measures: model certainty, data quality, or the likelihood the recommendation is correct. Then explain its limits and the uncertainty that remains, in plain language.

2

Evidence and action

Show the basis, match the workflow

Reveal contributing signals, source lineage, freshness, and a short explanation. High confidence streamlines the path while keeping evidence available. Low confidence invites review, comparison, or escalation. Either way the analyst can correct, give feedback, or override.

3

Calibration and behavior

Measure use, not display

Across comparable cases, an 80% score should prove correct about eight times out of ten. So the thing worth instrumenting is behavior: acceptance and overrides, ignored scores, errors and recovery. Watch for blind acceptance and habitual dismissal, because both mean the number is not doing its job.

The Why This Matters panel naming the technique, the confidence ring, and the recommended action together in one view
Score, evidence, and consequence in one panel
Design goal
Calibrated trust, not maximum trust

An analyst who trusts the assistant exactly as much as it deserves is the target. Analyst trust rose from 62% to 81%, which is the number I would defend, because it moved without the accuracy claim changing.

07 · Built with engineering

A confidence score is not a design decision or an engineering decision. It is both, or it is a number nobody should trust.

The percentage had to combine measurable signals from the response pipeline and then map that uncertainty onto product behavior. Neither discipline could specify it alone, so we ran five stages with a design and an engineering track through each one.

Design track Engineering track One number, shipped
01

Frame

Design

Define what analysts need to judge

Engineering

Identify measurable system signals

02

Instrument

Design

Specify evidence and state requirements

Engineering

Expose freshness, sources, and conflicts

03

Contract

Design

Map confidence to interface behavior

Engineering

Build the API and fallback logic

04

Validate

Design

Test interpretation and recovery

Engineering

Run calibration against reviewed cases

05

Monitor

Design

Study acceptance and overrides

Engineering

Track drift, latency, and failures

Decisions we made together

Four questions where a design answer without an engineering answer would have been fiction, and the reverse would have shipped a number with no meaning.

Score definition

What the percentage represents

Thresholds

When the experience should slow down

Human control

Which actions always require confirmation

Release criteria

When calibration supports production use

The artifacts were joint too: a confidence definition, a state model, an API contract, an evaluation plan, and a monitoring dashboard.

The contract, in practice

Design defined the experience contract. Engineering confirmed what the system could return reliably. This is the shape we settled on.

confidence-response.json Illustrative 
{
  "confidence": 0.89,
  "level": "high",
  "factors": [
    "MITRE mapping verified",
    "Three current sources"
  ],
  "limitations": [
    "Business criticality unavailable"
  ],
  "behavior": "review_and_confirm",
  "evidence_ids": [
    "asset-420", "mitre-T1134"
  ]
}
confidence Displays the calibrated percentage
factors Explains what supports the score
limitations Makes missing context visible
behavior Selects review or recovery state
evidence_ids Returns focus to source content

Illustrative shape, written for discussion. It does not reproduce proprietary production code.

08 · What I cut

Multi-step autonomous execution. The assistant could have chained patch, policy update, and notify with no human between them. I scoped it to one action, one confirmation, every time, trading speed for trust, on purpose.

Designing with AI

AI accelerated exploration, not judgment. I used it to generate alternative response structures, uncover failure cases, and turn promising directions into working interactions. Analyst research, security constraints, accessibility requirements, engineering feasibility, and my product judgment determined what moved forward.

  1. IllustrativeEarly open-ended responseFree prose. Slow to scan and hard to verify, as shown in What failed.
  2. IllustrativeStructures and failure casesAlternative response structures, checked against failure cases such as missing evidence and low confidence.
  3. ShippedEvidence-linked, confidence-awareThe interaction that moved forward after analyst research and security, accessibility, and engineering review. Try the recreation.

AI did not run the research or make the product decisions. The shipped design came from the team's work with analysts.

09 · Across the stack

I owned the interaction model end-to-end. Not a chat surface bolted onto dashboards, but how the AI reasons, grounds itself in evidence, and earns trust at each layer.

Analyst sees this layer
D
Tool / Agent
Proposes actions. A human confirms before anything runs.
C
Orchestration
Turns a question into cards, evidence and next steps.
B
Data
Grounds every claim in a tagged, linked source.
A
Model
Constrained by a schema and explicit rules.
Each layer builds on the one below
E
Performance
F
Safety
Four stacked layers, with performance and safety running through all of them
A · Model
Schema-constrained responses and rules-based behavior around model output, not open-ended generation.
Lower hallucination risk, faster scanning
B · Data
Every response evidence-linked, source-tagged, schema-bound.
78% evidence validation rate
C · Orchestration
Chat blended with structured cards and suggested follow-ups.
Question visual workflow
D · Tool / Agent
AI could filter, search, and map exposures, with every action traceable.
8% “AI Uncertain” rate
E · Performance
Streaming layouts and status messaging under real latency.
Responsive on large datasets
F · Safety
Confidence indicators, recovery paths, role-based gating.
Trust by explicit calibration

Seven patterns that travel

These came out of the Interpres work, and each holds up on any product surface where AI proposes, explains, or acts. They are reusable patterns from this project, not patterns I have deployed in other products.

PatternIn InterpresRule that travelsStatus
Context inheritanceThe dock inherits the active exposure, asset, or technique.A question never starts from nothing. The surface supplies the context.Shipped
Evidence and provenanceEvery claim is source-tagged, linked, and returns focus to the source.Every claim points to something the person can inspect.Shipped
Confidence and limitationsThe score is named, its basis shown, and its limits stated.Say what a number measures and what it misses.Shipped
Progress and latencyStreaming layouts and status messages under real latency.Say what the system is doing while it works.Shipped
Refusal and recoveryAn “AI uncertain” state explains the gap and offers another question.When it cannot answer, give a next step instead of a guess.Shipped
Human confirmationOne action, one confirmation, with the impact in the analyst’s terms.A consequential change stops and states what will change.Shipped
Result and traceabilityActions are traceable. The recreation shows an applied, logged state.Show what happened, where it is recorded, and how to recover.Concept extension

10 · Response

62 81%
Analyst trust
78%
Evidence validation
4 - 6
Follow-ups per chain
8%
AI uncertain rate

“Feels like a junior analyst who already knows our environment.”

Senior SOC Analyst

“For the first time, I can brief leadership using language they understand.”

CISO

Source: Mixpanel and OneStory analytics, measured during Beta.

In cybersecurity, ambiguity is risk. Design must reduce it.

The design system
Next case studyEmergency & threat intelligence platformRead the case study