Saraschandra KaranamHuman - AI Interactions Researcher · Enterprise AI
Research

Research

Selected research programs and ongoing experiments.

RESEARCH PROGRAMS

Sustained research programs exploring how people make decisions as technology becomes increasingly complex and intelligent.

PROGRAM 01·Cisco Systems·2025-26
Enterprise AI & Agentic Systems

"How do security professionals decide what to delegate to increasingly autonomous AI?"

Investigating how enterprise security professionals build trust, delegate decisions, and collaborate with increasingly autonomous AI systems during complex firewall management.

Focus Areas
Enterprise AIMixed MethodsTrust CalibrationAgentic Workflows
Impact:Prioritized list of agentic use cases and evaluated agentic touchpoints framework
Explore Program
Task complexityUser comfort with delegation to AIStrategic OpportunitiesQuick WinsWorkflow AWorkflow BWorkflow CWorkflow DWorkflow EWorkflow FWorkflow G
PROGRAM 02·Cisco Systems & MathWorks·2017-26
Complex Enterprise Workflows

"How can technology empower domain experts to orchestrate complex infrastructure?"

How to empower domain experts to orchestrate complex infrastructure using automation or by interacting with an abstracted representation of a system?

Focus Areas
Complex Enterprise WorkflowsMixed MethodsOperator ArchetypesCluster Analysis
Impact:Influenced product roadmaps for next-generation network/security management and model-based design systems
Explore Program
Research /EvaluationPurchaseStagingDay 0onboardingDay NconfigurationMonitoringTroubleshootingUpgrades
PROGRAM 03·Xerox Research Center India·2012-14
Human Computation & Crowdsourcing

"How can business owners decide which crowdsourcing platform to choose to meet enterprise SLAs?"

Benchmarking of crowdworker's performance mapping task quality and efficiency on crowdsourcing platforms leading to development of a recommendation engine - CrowdUtility

Focus Areas
Human ComputationCrowdsourcingTask QualityPerformance Optimization
Impact:Designed task assignment models achieving business SLA constraints with optimal cost
Explore Program
Task <Description ; QoS ; $ Budget>Digitization?Image Labeling?CROWD UTILITYPlatform StatisticsPlatform ModelsRecommendation Engine1234Cost/UnitAmazon MTurkCloud Factory

RESEARCH EXPERIMENTS

Explorations into how AI can augment the way research itself is conducted.

EXPERIMENT·2025-26

AI-Augmented Research

Exploring how AI can augment the way UX research is conducted.

Key Finding: The bottleneck shifts, it doesn't disappear. While the AI reduces tagging time, the human researcher must spend more time designing validation schemas and auditing outputs.
Explore Experiment
Search: Ask questions across feedback sourcesAsk me anything about customer feedback...SearchWhat do customers think about new navigation?Show me dashboard navigation issues in the last weekREACT Single Page AppSearch InterfaceSearch ExecutorAI Output SynthesizerMCP ClientOllama ClientHTTP / SSE / POSTOllama: Mistral 7BNode.js serverJSStdinStdoutFilesystem MCP serverNPS / CSATIntercom ConversationsSupport Tickets

ACADEMIC FOUNDATIONS

Earlier academic research that established the foundations for my work on information, cognition, and human-system interaction.

ACADEMIC FOUNDATION

Cognitive Modeling of Web Navigation & Information Search

2007–2011 · 2014–2017

IIIT Hyderabad & Utrecht University

Investigating how cognitive factors, expertise, age, and domain knowledge influence information search and web navigation.

Foundational Finding: Incorporating semantic information from pictures or individual differences with domain knowledge and age enhances the efficiency of cognitive models of web-navigation / information search.
Cognitive ModelingWeb-NavigationInformation SearchDomain KnowledgeAge Differences
Explore Research