Data Scientist 2
ZoomcarOwn and improve production pricing systems across base fares, product and flexibility recommendations, dynamic multipliers, asset controls, alerting and audited business changes.
CURIOUS LEARNER DATA SCIENTIST · PYTHON DEVELOPER
Archit Roy
I build production data-science systems where models meet products: pricing, behavioural prediction, generative AI, evaluation, APIs and the machinery that keeps them trustworthy.
02
About
I build production data-science systems at the point where models meet products: dynamic pricing, behavioural prediction, generative AI, evaluation, APIs and the operational machinery that keeps them trustworthy.
My default language is Python. I enjoy going from first-principles research and exploratory notebooks to data contracts, failure modes, observability and maintained services. I care as much about proving when a model should not act as I do about improving its score.
03
Experience
Data Scientist 1
ZoomcarBuilt and operated ML and generative-AI systems for booking intent, marketplace imagery, evidence-grounded support automation and analytical observability; designed experiments and production evaluation around them.
Data Science Intern
ZoomcarResearched twenty-five years of text-summarisation approaches and built DS-Nemo: an evaluated offline generation pipeline with low-latency APIs serving more than 200K daily requests.
04
Selected systems
Four featured systems lead the index. Filter by the capability you want to inspect.
Review summarisation
A review-summarisation system developed through classical NLP, fine-tuned transformers and LLMs, then productionised around evaluation and low-latency serving.
Associated with a 46% reduction in review clicks and a 12% reduction in checkout-session duration. The API handled 200K+ daily requests with an uncached p99 of 17 ms.READ CASE STUDY →
Booking-intent ranking
A production propensity system and controlled experiments that separated predictive ranking from intervention impact.
The top decile converted at about 3.16× baseline; experiments showed that a strong ranker does not guarantee a useful intervention.READ CASE STUDY →
Marketplace pricing operating system
Base fares, behavioural ranges, dynamic multipliers, product economics, interventions and change management treated as one ecosystem.
Made high-impact pricing changes faster, more explicit and easier to trace without a disruptive rewrite.READ CASE STUDY →
Guarded image enhancement
An end-to-end generative image pipeline with object-consistency checks, aesthetic scoring and marketplace experiments.
Produced directional CTR and ranking improvements across city experiments while filtering many harmful edits.READ CASE STUDY →
Outbound review voice agent
A policy-aware multilingual voice campaign for giving the platform's quiet majority of friction-free trips a fair chance to be represented in app-store reviews.
Produced a concise, versioned conversation specification and adversarial scenario matrix for the campaign.READ CASE STUDY →
Evidence-grounded support automation
An LLM-assisted support system improved through manual false-positive analysis, stakeholder adjudication and duplicate-safe processing.
Reached roughly 51% automated resolution across eligible flows at about 94% audited accuracy.READ CASE STUDY →
Pricing anomaly evaluation
A daily evaluator that answers a simple question: did a business metric genuinely move, or are we looking at normal variation or incomplete data?
Converted an exploratory workbook into a replay-tested daily service with explicit data-quality states.READ CASE STUDY →
Operational reporting platform
A common reporting layer that replaced scattered mailers with gated, previewable and consistent operational communication.
Created a reusable last mile for analytical products without duplicating their business logic.READ CASE STUDY →
Central DS–DE reliability alerts
A central reliability layer for execution health, infrastructure failures and database integrity across production data-science and data-engineering services.
Created a common operational safety net across the DS–DE production estate.READ CASE STUDY →05
Technical range
LANGUAGES
Python · SQL · C++
ML + STATS
XGBoost · Scikit-learn · SHAP · Pandas · NumPy · Experimentation
NLP + GENERATIVE AI
Gemini · Transformers · T5 · BART · PEGASUS · RAG · LLM evaluation
SYSTEMS
Flask · Gunicorn · SQLAlchemy · Docker · Kubernetes · Kafka
DATA + CLOUD
BigQuery · MySQL · GCP · Vertex AI · ETL pipelines
OBSERVABILITY
Langfuse · OpenTelemetry · New Relic · Last9 · Jenkins · ArgoCD
06
Education
Vellore Institute of Technology
B.Tech, Information Technology · CGPA 9.09 / 10.0
Kunskapsskolan
CBSE · Grade 12: 94% · Grade 10: 91%
AWS Certified Cloud Practitioner · Microsoft Power Platform App Maker Associate