KASHII UPDATEZ Everyday Student Requirements & Python Coding Tutorials by Python Kashi
JD #2687
Applications open Internship

India Runs on Voice Hackathon – Scaler x Wispr

Hackos

  • Pan India (online & on‑campus)
  • No salary – prize‑based participation (₹30 L total prize pool)
  • Fresher (0-1 Yrs)
  • Posted Today
HR
Recruiter contact
Recruiter / hiring contact
No public email in this post. Use a source-linked contact instead.

About the role

Scaler, a leading Indian ed‑tech platform that up‑skills software engineers, has partnered with Wispr Flow, a cutting‑edge voice‑to‑text AI, to launch a nation‑wide hackathon called “India Runs on Voice”. The event is designed to tap into India’s linguistic diversity and create voice‑first tools that can be used by billions of people across the country. Scaler’s mission is to democratise high‑quality tech education, while Wispr’s vision is to make voice the primary interaction mode for every digital product.

Together they provide participants with mentorship, master‑classes, and access to Wispr Flow’s API, enabling rapid prototyping of voice‑driven solutions.

The hackathon runs in three rounds from 5 Nov 2026 to 13 Dec 2026. Round 1 is an online warm‑up where participants submit ideas, prompts and impact statements using only their voice. Round 2 offers a unified problem statement; teams of up to three build a functional prototype, guided by a live master‑class from Wispr engineers.

The final round is an on‑campus showcase where shortlisted teams demo their solutions before Wispr leadership, AI founders and investors. The prize pool of ₹30 L is split among the top 15 teams, with the first‑place team receiving ₹7 L and a year’s subscription to Flow Pro for each member.

Key responsibilities for participants include

  • Identify a real‑world problem within one of the five theme tracks (civic services, healthcare, education, small business, everyday access).
  • Conduct rapid user research in regional languages to validate the problem.
  • Design voice‑first user flows that work hands‑free on low‑spec devices.
  • Implement the solution using Wispr Flow’s SDK, integrating speech‑to‑text, intent detection and language translation where needed.
  • Optimize the prototype for latency, accuracy and offline capability.
  • Prepare a compelling pitch deck and demo script entirely in voice.
  • Collaborate with teammates via voice‑only communication tools.
  • Iterate based on mentor feedback during the master‑class.
  • Present a live demo at the campus finale, answering technical and business questions.
  • Document the code and share open‑source contributions for community impact.

The tech stack revolves around Wispr Flow (speech‑to‑text, natural language understanding), any front‑end framework (React, Flutter, Android), backend services (Node.js, Python Flask), and cloud platforms (AWS, GCP). Participants gain hands‑on experience with voice AI, multilingual NLP, and rapid prototyping—skills highly sought after by product companies building inclusive tech.

Growth path: successful participants can be invited for internships or full‑time roles at Wispr, Scaler’s product teams, or partner startups looking for voice‑first innovators. The exposure to investors and AI leaders also opens doors to startup funding or mentorship.

Why join? This is a unique chance to build for a billion users, work in any Indian language, and receive mentorship from industry experts—all without any entry fee. Winners walk away with cash prizes, a year of premium Flow Pro, an international trip to Singapore, and direct interview opportunities with Wispr. Even non‑winners gain a portfolio‑ready voice‑AI project and a network of like‑minded builders across India.

Eligibility & process

Who can apply

Open to all Indian students (undergraduate, postgraduate) and working professionals. No specific degree or branch requirement. Candidates from any batch year may apply. Minimum 60% academic record preferred but not mandatory. No backlog restrictions; participants must be legally eligible to work in India.

Skills & stack

Voice Ai Speech‑To‑Text Natural Language Processing Python JavaScript Mobile App Development UI Ux Design For Voice

Selection process

  1. Round 1: Idea, Prompt and Impact submissions (voice‑only)
  2. Round 2: Build prototype and submit by 29 Nov (online master‑class included)
  3. Round 3: On‑campus finale demo on 13 Dec (shortlisted teams only)

Prep kit

19 resources
Study material 4
Company-specific guides 4
Likely coding questions 4
Round-wise strategy 4
Round 1: Online Assessment (OA) Aptitude, Quantitative Logic & 2 Coding Problems Focus on accuracy and speed. Practice arrays, strings, and standard arithmetic puzzles.
Round 2: Technical Interview 1 Core Tech Stack (Voice Ai, Speech‑To‑Text, Natural Language Processing, Python, JavaScript, Mobile App Development, UI, Ux Design For Voice) & Live Code Tracing Explain your thought process aloud. Analyze time & space complexities before coding.
Round 3: System Design & Problem Solving Database Schemas, APIs & Architecture Basics Clarify edge cases, diagram schemas cleanly, and discuss scalability trade-offs.
Round 4: HR & Cultural Fit Discussion Hackos Core Values, Learning Agility & Offer Terms Demonstrate passion, strong communication, and readiness for full-time collaboration.
Technical questions 4
What is the bias-variance tradeoff and how do you prevent overfitting? Answer
Model answer: High bias leads to underfitting (oversimplified model), high variance leads to overfitting (captures noise). Mitigate using L1/L2 Regularization, Dropout, Cross-Validation, and data augmentation.
Explain the difference between Precision, Recall, and F1-Score. Answer
Model answer: Precision = TP / (TP + FP) (correctness of positive predictions). Recall = TP / (TP + FN) (coverage of actual positives). F1-Score is the harmonic mean of Precision and Recall.
How does Gradient Descent work and what is the role of Learning Rate? Answer
Model answer: It optimizes loss functions by iteratively moving weights in the direction of negative gradient. A large learning rate may overshoot the minimum; a small rate causes slow convergence.
What is the difference between Supervised, Unsupervised, and Self-Supervised learning? Answer
Model answer: Supervised uses labeled data (X -> y). Unsupervised finds hidden patterns in unlabeled data (clustering/PCA). Self-supervised generates labels from input data (e.g. masked language modeling in BERT/Transformers).
Behavioural & HR questions 3
Why do you want to join Hackos as a India Runs on Voice Hackathon – Scaler x Wispr?
Highlight Hackos's market reputation, recent tech innovations, and how your skills in Voice Ai directly solve their team's objectives.
Describe a challenging bug or academic project roadblock and how you resolved it.
Use the STAR method: Situation (project context), Task (what needed solving), Action (specific tools/logic applied), Result (quantifiable positive outcome).
How do you handle strict deadlines or sudden scope changes?
Explain your prioritization strategy, proactive communication with mentors/peers, and agile mindset.
Hackos No salary – prize‑based participation (₹30 L total prize pool)
Recruiter Resume Apply
Chat Chat with Kashii