Senior or staff ML / AI scientist roles (LLMs, agentic AI, ads and recommendation, risk modelling, time series) at product companies, funds or AI startups. Remote-first preferred; open to Bengaluru, Hyderabad, Pune or Gurgaon. 2-month notice, negotiable.

Nikhil Mishra
@nikhilmishra5
Senior Data Scientist at Pattern. Kaggle Competitions Grandmaster (7 golds). LLMs, agentic AI, ads bidding, lending risk. Top 3 in 40+ ML contests.
What I'm looking for
At Ads Pattern®, I integrated product targeting into Destiny for Amazon Sponsored Products, adding ASIN- and category-level target selection with optimal bids. The change replaced manual, history-based target selection and improved spend by 20%.
I also launched a Sponsored Brands keyword-targeting module and improved the conversion-rate and CPC prediction models behind real-time ad-spend allocation. By optimizing the keyword-targeting pipeline, I cut its runtime from three hours to 25 minutes.
Before Ads Pattern®, I built lending models and production ML systems at OkCredit and KarmaLife, and worked on fraud detection, document-based investment reports, and intent classification at H2O.ai. I’m also a Kaggle Competitions GrandMaster, with top-three finishes in more than 40 data science competitions.
Experience
Work history, roles, and key accomplishments
Sr. Data Scientist
Ads Pattern
Oct 2025 - Present (1 year 1 month)
Integrated a product targeting system into Destiny for Amazon Sponsored Products, an ML predictor and optimizer that runs automated bidding under a dual-constrained objective of maximizing both ROAS and spend, serving 13 Amazon marketplaces, 250+ brands, and 40K+ products. Launched a Sponsored Brands keyword targeting module and improved conversion-rate and CPC prediction models.
Sr. AI Scientist
Agentic AI NeuralHQ
Mar 2025 - Sep 2025 (6 months)
Architected a scalable hierarchical Agentic AI framework for customer conversations, achieving 95% SOP adherence across 1,000+ documents with avg response latency <4 seconds. Built Auto-Test & Auto-Eval pipelines to auto-generate mock API data and user scenarios, reducing end-to-end evaluation cycle to under 2 minutes.
Helped large banking clients develop fraud detection and early defaulter identification models using ensemble methods and graph-based features, reducing their credit loss exposure by ~18%. Built an LLM-powered system that generates investment reports automatically from 500+ page documents.
Sr. Data Scientist
OkCredit
Dec 2021 - Feb 2024 (2 years 2 months)
Single-handedly built the loan pre-approval system for 2.6M merchants using LightGBM models on alternate data, in-app usage signals, and network-interaction features, achieving 0.70 AUC with NPA held under 3% on disbursed loans. Owned the entire pipeline end to end on Google Cloud, from a merchant aggregator to production serving, orchestrated with Airflow on GKE.
Data Scientist
KarmaLife
Dec 2020 - Nov 2021 (11 months)
Built the ML risk model for weekly salary-advance loans to gig-economy workers, deciding who qualifies for a loan and how much, using LightGBM on attendance, loan repayment history, and user metadata; cut loan default rates by 20%. Designed automated ETL pipelines on AWS and Apache Superset dashboards to monitor the loan portfolio, repayments, and defaults in real time.
Education
Degrees, certifications, and relevant coursework
PESIT
Bachelor of Engineering, Computer Science and Engineering
2016 - 2020
Grade: 7.5 CGPA
Bachelor of Engineering in Computer Science and Engineering from PESIT, Bangalore, with a CGPA of 7.5.
Delhi Public School, Siliguri
CBSE XII, Science
2014 - 2016
Grade: 88.8%
Completed CBSE XII Science with 88.8%.
Availability
Location
Authorized to work in
Salary expectations
Social media
Skills
Interested in hiring Nikhil?
You can contact Nikhil and 600K+ other talented remote workers on Himalayas.
Message NikhilGet matched with your dream remote job
Sign up now and join over 1,000,000+ remote workers who receive personalized job alerts, curated job matches, and more for free!
