Hello,
I'm Samyak
/suh · myuhk/
I'm a cynical developer. I build high-performance, cloud-native distributed systems and design them around the assumption that things break, because they do. I take products from 0 → 1 and evolve them into production-grade platforms, and I enjoy the unglamorous half: the retries, the recovery paths, and the failure modes nobody plans for.
Day job
Currently, I am a Software Engineer at Outengine, building B2B data infrastructure: a lead-search engine over 800M+ profiles, a semantic company lookalike engine, and a distributed enrichment system sustaining 7.5M+ domains/day. Before that, I was a Research Intern at IIT Delhi, building secure backend modules for verified data exchange, and an ML Intern at Liberin Technologies, developing data security tools using computer vision. I have participated in multiple hackathons, creating innovative solutions and earning recognition for my projects.
Out of office
When I'm not building things, I love traveling, playing video games, baking, cooking, painting, and capturing moments through photography. I'm a natural puzzler with a passion for trivia and crosswords, maybe even to a concerning degree! If I've piqued your interest, feel free to reach out!
How I Work
Structure in ambiguity
Most problems arrive underspecified, and waiting for someone else to specify them is how a quarter disappears. I would rather map the thing early, write down what is actually being decided, and be wrong while it is still cheap to fix.
Storytelling
I love taking complicated systems (schemas, retrieval pipelines, crawl waterfalls) and turning them into narratives the rest of the team can reason about and build on.
Scale as a constraint
I design for the size the data will actually be. Index tuning, skip indexes, queue backpressure and checkpointed resumes are decisions I make up front, not patches I bolt on once things start falling over.
0 → 1, then production
I like being there for both halves: shipping the first working version fast, then doing the unglamorous work (idempotency, rollback handling, health monitoring) that turns it into something people can depend on.
Engineering Principles
Learned the expensive way, at 3am, in production.
Design for failures instead of assuming success.
The happy path is the least interesting path, because it's the one that almost never runs. Disks fill, DNS lies, the upstream API returns 200 with an empty body. I write the code that assumes all of that already happened.
Build systems that degrade gracefully under load.
Under enough load something has to give. If I haven't decided what, the system decides for me, usually at the worst possible moment, and usually everything at once.
Prioritize observability, retries and recovery mechanisms.
A system you can't see into isn't healthy, it's just quiet. Retries, checkpoints and auto-resume exist because I'd rather the job recover itself than page me to press the same button it could have pressed.
Optimize for throughput without sacrificing reliability.
Benchmarks are easy to win if you're willing to be wrong quickly. Fast and wrong is still wrong. It just arrives sooner and in higher volume.
Automate everything that can be automated.
Anything done by hand twice gets done wrong the third time, and nobody will remember which step was skipped. Scripts don't get bored, don't get tired, and don't improvise at 3am.
Experience
Sept 2025 - Present
Outengine
Software Engineer
- Built a B2B lead-search engine over 800M+ people profiles and 62M companies, designing the schema and index strategy to serve 40+ filter dimensions in 700ms-1s.
- Built a semantic company lookalike engine over 25M+ records, combining dense, pseudo-relevance-feedback and keyword retrieval with calibrated fusion and a 790M-edge talent-flow competitor graph, tuned via a 200-config benchmark.
- Architected a distributed data-enrichment system for a 340M+ domain universe, sustaining 7.5M+ domains/day across 13 servers at 84-89% success via a 6-tier crawl waterfall with stealth browser rendering.
- Engineered an asynchronous email-enrichment pipeline with SMTP verification, credit accounting, real-time progress streaming and checkpoint-based auto-resume, served behind a mutually authenticated data gateway.
- Built the email infrastructure with automated server, DNS and mailbox provisioning, plus reliable backend automation: REST APIs, idempotent workflows, rollback handling and server health monitoring.
May 2025 - May 2026
Indian Institute of Technology Delhi (IIT Delhi)
Research Intern
Advised by Prof. Kaustubh Beedkar and Prof. Amitabha Bagchi.
- Designed secure backend modules for verified data exchange, deployed on AWS EC2 with S3 storage and SES-based email verification, including agreement management workflows for safe and compliant data requests.
- Designed and implemented a secure user invitation workflow with role-based access control (RBAC) to manage permissions across organizations.
July 2024 - October 2024
Liberin Technologies
ML Intern
- Engineered a Smart ID Masking application using Python, Flask and YOLO, enabling secure card-type detection and sensitive information protection.
- Developed a selective masking feature for controlling visibility of personal identifiers, applying computer vision techniques for reliable detection and anonymization across diverse ID card formats.
Skills
Languages
AI, ML & Retrieval
Backend & APIs
Databases & Vector Search
Data Engineering & Scraping
DevOps & Cloud
Certifications
2025
Applied AI
WorldQuant University
Certified proficiency in deep learning for computer vision, including skills in neural networks, image classification, object detection, and generative AI through hands-on projects.
2024
Applied Data Science
WorldQuant University
Certified proficiency in completing end-to-end data science projects, including data exploration, machine learning model development, ETL pipelines, and data visualization.
2023
Data Science and Machine Learning Professional Certificate
IBM
Acquired practical knowledge in data preprocessing, visualization, and machine learning techniques.
Extra-Curricular Activities
Stacks Hacker House - Goa Edition
Selected for the Stacks Hacker House in Goa, a residency for builders shipping on Stacks, the smart-contract layer for Bitcoin.
Build & Brew - Espresso Global Hackathon Bounty Winner
Won a bounty prize in the Caffeinate and Code track for innovative blockchain development work during the Build and Brew Hackathon powered by Espresso.
Stellar Builders Residency - Kasol Edition
Selected for an exclusive 4-day immersive experience designed for blockchain builders, Web3 innovators, and ecosystem pioneers to build real-world solutions using the Stellar network in the Heart of the Himalayas (27-30 April 2025).
Winter School on Data Systems - IIT Delhi
Learned distributed data processing, stream processing with Spark & Flink, and hands-on labs on fault-tolerant systems.
Summer Analytics 2024 (IIT Guwahati) - Certificate of Merit
Ranked in the top 25% in the Summer Analytics 2024 Data Science & ML course by IIT Guwahati’s Consulting & Analytics Club.
Education
Nov 2022 - May 2026
Guru Gobind Singh Indraprastha University (IPU)
B.Tech in Artificial Intelligence and Data Science
Specialized in Machine Learning, Data Science, and Software Development, with hands-on experience in AI-powered systems.
Relevant coursework: Data Science, Machine Learning, Probability, Linear Algebra, AI Principles, DBMS, Software Engineering.
Get in touch
Let's talk
Have a project in mind? Looking to collaborate or work together? Book a call, or send a message, whichever is easier.
15 or 30 minutes. Pick a slot that works for you.