Vasu Sadariya profile photo 1 on Tal

Vasu Sadariya

Founding AI Engineer @CareerPassport.ai | LLM & AI Integration | Full Stack Dev | GWoC Winner · DotSlash Finalist | Final Year ECE | SVNIT’27

Surat

2+ years of work experience

Stealth Startup
May 20262 moJul 2026

Bengaluru

As one of the first engineers on the AI surfaces of the product, spent two months building the LLM-driven pipelines the platform now runs on: • Owned the job-brief LLM extraction pipeline end-to-end — added retries with backoff, structured production logging, and boundary-level coercion so messy real-world model output (arrays coming back as strings, nulls, inconsistent shapes) stopped silently breaking downstream code. • Designed and built an AI video-analysis pipeline from scratch: schema, Zod-typed contracts, a hand-rolled REST client for a third-party AI video provider (the API wasn't in their official SDK), and background-job orchestration running summary/key-moment/Q&A generation in parallel with idempotent, replay-safe retries. • Root-caused a production-only incident — AI analysis taking 3x longer in prod than locally — to a silent webhook timeout eating 94% of the time budget, by reading the actual production trace instead of guessing. Replaced it with a bounded self-poll pattern and cut a related pipeline's worst case from 9 minutes to 3. • Shipped voice-answer recording for candidates: mic capture, parallel transcription, automatic dedup on re-record, signed playback for recruiters. • Closed out the internship auditing real per-request LLM costs and rate limits across every AI-billed flow on the platform (job-brief generation, resume parsing, trip generation) to design a safe load-testing capability for concurrent AI-generation traffic under a morning-rush scenario. Also picked up a good chunk of core product work outside the AI surfaces along the way — 61 production changes shipped in total across two frontend apps and two backend services in two months.
MINDBEND, NIT Surat
Feb 20266 mosJul 2026

MINDBEND, NIT Surat
Aug 202511 mosJun 2026

Surat

Cutting Edge Visionaries
Jun 20251 yr 2 mosJul 2026

Surat, Gujarat, India

In my role as Chief Executive of the student chapter at Nit Surat, I spearheaded initiatives to advance knowledge in fintech through engaging workshops and events. By leveraging the resources of our club, we were able to provide valuable networking opportunities and industry insights to our members, fostering a culture of continuous learning and growth.
May 20253 mosJul 2025

Delhi, India

During the internship period i work as a Core Developer & DevOps lead on a three-member team: designed Docker and GitHub Actions CI/CD pipelines, built responsive front-end UIs, and implemented robust back-end services.Automated deployments, reducing cycle time by 70%. Single handedly Deployed whole frontend and backend in AWS in Ec2 using NGINX also do certificate management
DSC NIT Surat
Nov 20241 yr 9 mosJul 2026

Surat

Work as a junior Developer
Cutting Edge Visionaries
Aug 202411 mosJun 2025

India

MINDBEND, NIT Surat
May 20241 yr 11 mosMar 2026

Surat, Gujarat, India

Major Duties and serving my experties in Technical And Sponsorship committee

Vasu’s AI Stack

anthropic apis
anthropic apis
claude code
claude code
claude
claude
codex
codex
langfuse
langfuse
n8n
n8n

Education

Sardar Vallabhbhai National Institute of Technology, Surat

Bachelor of Technology - BTech, Electronic and Communications Engineering

Sardar Vallabhbhai National Institute of Technology, Surat

20234 yrs2027
Dholakiya schools rajkot-gondal

X, GSEB

Dholakiya schools rajkot-gondal

Join Vasu and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Vasu Sadariya on Tal

Headline

Founding AI Engineer @CareerPassport.ai | LLM & AI Integration | Full Stack Dev | GWoC Winner · DotSlash Finalist | Final Year ECE | SVNIT’27

Location

Surat

AI stack

anthropic apis

claude code

claude

codex

langfuse

n8n

Work experience

AI Engineer Intern at Stealth Startup, May 2026 - Jul 2026 · 2 mo, Bengaluru, As one of the first engineers on the AI surfaces of the product, spent two months building the LLM-driven pipelines the platform now runs on: • Owned the job-brief LLM extraction pipeline end-to-end — added retries with backoff, structured production logging, and boundary-level coercion so messy real-world model output (arrays coming back as strings, nulls, inconsistent shapes) stopped silently breaking downstream code. • Designed and built an AI video-analysis pipeline from scratch: schema, Zod-typed contracts, a hand-rolled REST client for a third-party AI video provider (the API wasn't in their official SDK), and background-job orchestration running summary/key-moment/Q&A generation in parallel with idempotent, replay-safe retries. • Root-caused a production-only incident — AI analysis taking 3x longer in prod than locally — to a silent webhook timeout eating 94% of the time budget, by reading the actual production trace instead of guessing. Replaced it with a bounded self-poll pattern and cut a related pipeline's worst case from 9 minutes to 3. • Shipped voice-answer recording for candidates: mic capture, parallel transcription, automatic dedup on re-record, signed playback for recruiters. • Closed out the internship auditing real per-request LLM costs and rate limits across every AI-billed flow on the platform (job-brief generation, resume parsing, trip generation) to design a safe load-testing capability for concurrent AI-generation traffic under a morning-rush scenario. Also picked up a good chunk of core product work outside the AI surfaces along the way — 61 production changes shipped in total across two frontend apps and two backend services in two months.

Joint Co-curricular affairs Secretary at MINDBEND, NIT Surat, Feb 2026 - Jul 2026 · 6 mos

Manager at MINDBEND, NIT Surat, Aug 2025 - Jun 2026 · 11 mos, Surat

Chief Executive at Cutting Edge Visionaries, Jun 2025 - Jul 2026 · 1 yr 2 mos, Surat, Gujarat, India, In my role as Chief Executive of the student chapter at Nit Surat, I spearheaded initiatives to advance knowledge in fintech through engaging workshops and events. By leveraging the resources of our club, we were able to provide valuable networking opportunities and industry insights to our members, fostering a culture of continuous learning and growth.

Software Engineer Intern at Qwykli, May 2025 - Jul 2025 · 3 mos, Delhi, India, During the internship period i work as a Core Developer & DevOps lead on a three-member team: designed Docker and GitHub Actions CI/CD pipelines, built responsive front-end UIs, and implemented robust back-end services.Automated deployments, reducing cycle time by 70%. Single handedly Deployed whole frontend and backend in AWS in Ec2 using NGINX also do certificate management

Junior Developer at DSC NIT Surat, Nov 2024 - Jul 2026 · 1 yr 9 mos, Surat, Work as a junior Developer

Executive at Cutting Edge Visionaries, Aug 2024 - Jun 2025 · 11 mos, India

Co Head at MINDBEND, NIT Surat, May 2024 - Mar 2026 · 1 yr 11 mos, Surat, Gujarat, India, Major Duties and serving my experties in Technical And Sponsorship committee

Education

Bachelor of Technology - BTech, Electronic and Communications Engineering, Sardar Vallabhbhai National Institute of Technology, Surat

X, GSEB, Dholakiya schools rajkot-gondal

Public links

github: https://github.com/vasusadariya

leetcode: https://leetcode.com/Vasu3007

vasusadariya.dev: https://www.vasusadariya.dev/