Tiyasha Khan

Associate Consultant

Bengaluru, Karnataka, India6 yrs 6 mos experience
AI ML PractitionerAI Enabled

Key Highlights

  • Led AI transformation for scalable workflows.
  • Developed predictive models driving client revenue.
  • Created automated reporting systems for data accuracy.
Stackforce AI infers this person is a Data Science expert in MarTech and AdTech with strong analytical capabilities.

Contact

Skills

Core Skills

Data ScienceMachine LearningData Analysis

Other Skills

AI transformationModel Context ProtocolLangChain toolsGoogle Ads APIMeta APIsAI-reporting systemdata accuracyautomated reportingLLM research frameworkmodel testingGoogle’s Meridian MMMN-gram NLPMarkov chain attribution analysisclient-centric modelling solutionsDatabricks

Experience

6 yrs 6 mos
Total Experience
4 yrs 6 mos
Average Tenure
1 yr 11 mos
Current Experience

Rebid

Associate Manager - Data Science

Jun 2024Present · 1 yr 11 mos

  • Led the Data Science team through an extensive AI transformation, scaling the Model Context Protocol to 20+ AI chat workflows using 12+ LangChain tools integrated with Google Ads and Meta APIs. Achieved eight seamless conversation threads at a time, extended browser timeouts from 60 seconds to 15 minutes, and implemented zero-downtime architecture.
  • Designed a modular AI-reporting system that replaced legacy pipelines with native integrations, ensuring 100% data accuracy and zero-redeployment client onboarding. Deployed 10+ automated reporting systems generating 50+ reports monthly.
  • Launched the AI Summary feature—now receiving ~3,500 monthly hits—providing dashboard insights and automated alert emailers. Introduced HTML-guardrail prompt engineering.
  • Established a bi-monthly LLM research framework, testing 25+ models (GPT‑4/4.1‑mini, Claude, Deepseek, Grok, Gemini) for latency, accuracy, and cost efficiency. Experimented with Google’s Meridian MMM, channel‑level propensity modelling, N‑gram NLP, and Markov chain attribution analysis.
  • Engineered 7+ tailored client‑centric modelling solutions, including PDP recommenders (cosine similarity), ATC Apriori basket analysis, hybrid PLP models (NMF + LSTM), lag‑difference MMM for daily budgets, custom bidding (126% click uplift), Intent‑to‑Purchase models, ad fatigue logic, and offline recommendation engines.
  • Drove infrastructure optimisation by onboarding Databricks and PySpark workflows for maintainability. Built cross‑team collaboration loops among CS, Dev, and DevOps to ensure continuous product improvement.
AI transformationModel Context ProtocolLangChain toolsGoogle Ads APIMeta APIsAI-reporting system+12

Miq

3 roles

Senior Data Scientist

Promoted

Feb 2022May 2024 · 2 yrs 3 mos

  • Led stakeholder collaborations to address client challenges, overseeing ML algorithm
  • development, driving revenue growth for MiQ through client.
  • Developed and deployed various analytical solutions such as predictive modelling using
  • XGBoost and A/B testing strategies to reach high potential buyers of client.
  • Developed a new targeting strategy using look-a-like modelling, by segmenting potential users’
  • behaviour from Australian Linear TV data.
  • Developed a recommendation engine to enhance tailored targeting for clients for
  • entertainment clients.
  • Consult and engage with stakeholders to assess the business problems, develop roadmaps,
  • and validate performance of models to support clients in various verticals such as Healthcare,
  • Insurance and Government Ad agencies.
ML algorithm developmentpredictive modellingA/B testinglook-a-like modellingrecommendation enginestakeholder engagement+2

Senior Business Analyst

Promoted

Jan 2021Feb 2022 · 1 yr 1 mo

  • Automated repetitive insights and reporting tasks, reducing monthly hours by 10.
  • Deployed end- to- end machine learning models trained on historic ad campaign budget data
  • by incorporating Automatic Bid Optimization, resulting in a 21% reduction in budget
  • overspending.
  • Implemented an end- to- end ML model trained on user website behaviour, employing Custom
  • Bidding with propensity modelling, resulting in a 1.5x increase in conversions while reducing
  • CPM costs.
  • Leveraged analytical solutions to drive strategic engagements with three or more clients within
  • the EMEA market, aimed at reducing churn.
machine learning modelsAutomatic Bid Optimizationpropensity modellinganalytical solutionsData Science

Business Analyst

Sep 2019Dec 2020 · 1 yr 3 mos

  • Developed insights for 50+ advertisers across diverse verticals, encompassing banks,
  • automobile, retail, food, and tourism, leading to enhanced client engagement and increased
  • spending with MiQ.
  • Created 5 client- facing analytical dashboards utilizing big data and viz tools like Tableau.
  • Assessed the advertising analytics landscape to facilitate the successful launch of the
  • business in a new market—Middle East, North Africa, and Turkey (MENAT). This involved
  • thorough requirement gathering, data onboarding, productionalizing, and strategy development
  • for various advertisers.
analytical dashboardsbig data toolsrequirement gatheringdata onboardingData Analysis

Ola (ani technologies pvt. ltd)

Intern - Decision Science

Jul 2019Sep 2019 · 2 mos · Bengaluru Area, India

  • Worked in the Underwriting team of Ola Financial Services.
  • Projects/Tasks:
  • 1. External Merchant Tracker: Created tool for automating external merchant tracking for Ola Money Post-paid, captured the week on week performances of the merchants.
  • 2. External Merchant Analysis: Created variables and modeled using R to obtain probability of bad for underwriting of customers in external merchants.
  • 3. Customer Acceptance Behavior Analysis: Created behavioral variables to track customer acceptance on Ola Money Post-paid platform.
automating trackingR programmingcustomer behavior analysis

Accenture in india

Functional and Industry Analyst- Intern

May 2018Jul 2018 · 2 mos · Bengaluru Area, India

  • Worked in the Functional and Industry Division of Accenture Digital.
  • Project/Task : Offer Price Optimization for Customer Retention
  • OTT business is a dynamic environment where customer retention is key for companies as customers can cancel their subscription at any point of time. The project involved predicting the probable reason for cancellation of any given product and add an additional predictive layer to decide on probable offer related solutions. Models were developed in R using Machine Learning and Artificial Neural Network algorithms. Built a prototype app in R-Shiny to showcase how retention can be increased by the help of ensemble models and interactive customer feedback by offering an optimized price based on the feedback provided.
offer price optimizationpredictive modelingR-Shiny

Education

Madras School of Economics

Master's degree — Financial Economics

Jan 2017Jan 2019

St. Xavier's College (Autonomous), Kolkata

Bachelor of Science - BSc. (Honours) — Statistics

Jan 2014Jan 2017

St Joseph's Convent

Higher Secondary — Science

Stackforce found 100+ more professionals with Data Science & Machine Learning

Explore similar profiles based on matching skills and experience