Devansh Jain profile photo 1 on Tal

Devansh Jain

SDE-1 @ Dream11 | Mathematics & Scientific Computing, Machine Learning

Mumbai

1+ year of work experience

Dream11
Jul 20251 yr 2 mosPresent

Mumbai, Maharashtra, India

Pulse — Mobile Observability Platform •Built core PulseKit iOS SDK features in Swift/OpenTelemetry — session replay, heatmaps, crash detection — now live at 250K+ DAU ingesting 20M+ events/day. •Designed a Kafka → ClickHouse pipeline that pre-aggregates millions of tap events into daily heatmap bins. •Shipped an AI diagnostics dashboard (React + Gemini) that ranks 9 issue types by user impact and writes root-cause narratives. Raven — Customer Engagement Platform •Built a server-driven UI engine in React Native that renders tooltips, popups, and bottom sheets from backend rules — no app release needed. •Designed a state-machine DSL (Vert.x + Aerospike) to model user journeys with frequency capping and event-based triggers. •Added multi-tenant OAuth 2.0/OIDC with PKCE and passwordless OTP across SDK endpoints. Shipped to Dream11's 15M+ DAU app; first rollout drove 10K+ monthly engaged users.
neurIOT Labs
May 20243 mosJul 2024

Gurugram, Haryana, India

• Developed summarization techniques for doctors’ tweets, enhancing the communication of medical information. • Built an NLP pipeline to process and summarize over 10,000 medical tweets, extracting key insights. • Achieved 90.4% prediction accuracy in a disease prediction model using semantic analysis of patient descriptions.
Indian Institute of Technology, Kanpur
May 20243 mosJul 2024

Kanpur, Uttar Pradesh, India

𝐌𝐞𝐧𝐭𝐨𝐫: Professor Raghvi Garg (Department of Economics, IIT Kanpur) • Conducted in-depth investigations into decision-making patterns to identify strategic behavioral differences. • Utilized the Kolmogorov-Smirnov test to enhance statistical robustness and improve decision accuracy. • Analyzed over 4,500 data points, achieving 80% accuracy in predicting behavioral responses.

Devansh’s AI Stack

opencode
opencode
claudecode
claudecode

Education

Indian Institute of Technology, Kanpur

Mathematics & scientific computing, Minor:- Machine Learning

Indian Institute of Technology, Kanpur

20214 yrs2025

Join Devansh and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Devansh Jain on Tal

Headline

SDE-1 @ Dream11 | Mathematics & Scientific Computing, Machine Learning

About

Full-stack engineer working across native iOS, react native SDKs, Vert.x backends, and React dashboards. I worked on observability and engagement platforms at scale — from instrumenting telemetry on-device to aggregating millions of events in ClickHouse. Strong mathematical foundation from IIT Kanpur, with hands-on ML and distributed-systems experience.

Location

Mumbai

AI stack

opencode

claudecode

Work experience

SDE-1 at Dream11, Jul 2025 - Present · 1 yr 2 mos, Mumbai, Maharashtra, India, Pulse — Mobile Observability Platform •Built core PulseKit iOS SDK features in Swift/OpenTelemetry — session replay, heatmaps, crash detection — now live at 250K+ DAU ingesting 20M+ events/day. •Designed a Kafka → ClickHouse pipeline that pre-aggregates millions of tap events into daily heatmap bins. •Shipped an AI diagnostics dashboard (React + Gemini) that ranks 9 issue types by user impact and writes root-cause narratives. Raven — Customer Engagement Platform •Built a server-driven UI engine in React Native that renders tooltips, popups, and bottom sheets from backend rules — no app release needed. •Designed a state-machine DSL (Vert.x + Aerospike) to model user journeys with frequency capping and event-based triggers. •Added multi-tenant OAuth 2.0/OIDC with PKCE and passwordless OTP across SDK endpoints. Shipped to Dream11's 15M+ DAU app; first rollout drove 10K+ monthly engaged users.

Data Science Intern at neurIOT Labs, May 2024 - Jul 2024 · 3 mos, Gurugram, Haryana, India, • Developed summarization techniques for doctors’ tweets, enhancing the communication of medical information. • Built an NLP pipeline to process and summarize over 10,000 medical tweets, extracting key insights. • Achieved 90.4% prediction accuracy in a disease prediction model using semantic analysis of patient descriptions.

Research Assistant at Indian Institute of Technology, Kanpur, May 2024 - Jul 2024 · 3 mos, Kanpur, Uttar Pradesh, India, 𝐌𝐞𝐧𝐭𝐨𝐫: Professor Raghvi Garg (Department of Economics, IIT Kanpur) • Conducted in-depth investigations into decision-making patterns to identify strategic behavioral differences. • Utilized the Kolmogorov-Smirnov test to enhance statistical robustness and improve decision accuracy. • Analyzed over 4,500 data points, achieving 80% accuracy in predicting behavioral responses.

Education

Mathematics & scientific computing, Minor:- Machine Learning, Indian Institute of Technology, Kanpur

Devansh Jain - SDE-1 in Mumbai | Tal