Aditya Singh
Open to work0 years experienceSoftware Engineer
Computer Science and Engineering Student | Software Engineering Intern
Standing Rank
Rank Not Available
Developer Badges
No badges earned yet
Skills & Technologies
24 skillsWork Experience
Software Engineering Intern
Eli Lilly and Company
- Built and shipped full-stack features (Next.js, React, TypeScript, PostgreSQL, Redis) across an enterprise assistant platform serving 200 internal employees to create and share domain-specific AI assistants.
- Integrated an external knowledge source (Veeva Vault) via the Model Context Protocol with SSE streaming, and built admin governance tooling for model whitelisting and per-assistant access control.
- Resolved the majority of critical/high frontend security vulnerabilities through dependency auditing and npm override strategies.
Research Intern
International Centre for Applied Sciences (ICAS)
- Built a hierarchical multi-head EfficientNet-B0 classification pipeline on DERM12345 (12,345 dermoscopic images, 40 subclasses), achieving 98.32% binary accuracy and 0.994 AUC — establishing early benchmarks on a dataset published in Scientific Data (Nature, 2024) with no prior classification baselines.
- Conducted a 6-model systematic study of loss functions (Dice, Tversky , Focal Tversky) on a 3D U-Net for BraTS 2020 brain tumor segmentation, achieving Dice scores of 0.77 (Tumor Core) and 0.75 (Enhancing Tumor) on 64³ voxel patches.
Software Engineer Intern
Recursive Zero
- Migrated multiple client websites from vanilla HTML/JS to Astro.js, cutting JavaScript bundle size by∼50% and improving Core Web Vitals and SEO across all projects.
Projects
HNSW Vector Search Engine
- Implemented HNSW approximate nearest-neighbour search from the original paper (Malkov & Yashunin, 2016); recall curve matches hnswlib/FAISS (95.6% recall@10), with the∼70× QPS gap attributable to pure Python vs. their SIMD-optimized C++.
- Built a reproducible benchmark harness measuring recall-vs-throughput on SIFT-1M set against two industrial C++ libraries, verifying correctness against an exact brute-force baseline.
- Wrapped the index in a FastAPI service with save/load persistence and a semantic-search demo.
Distributed Message Broker (Kafka-style)
- Built an append-only segmented log with per-write fsync durability , verified by crash-recovery tests (kill -9 mid-write→ zero data loss on restart).
- Designed a custom binary TCP protocol with length-prefixed framing, serving concurrent clients via a goroutine-per-connection architecture.
- Implemented Kafka’s StickyAssignor rebalancing algorithm and heartbeat-based failure detection for consumer group coordination, verified by unit tests.
Self-Healing RAG
- Built a corrective RAG pipeline that validates its own answers via LLM-as-judge and automatically retries with alternative retrieval strategies when confidence drops below threshold.
- Designed graceful degradation: surfaces partial answers with identified knowledge gaps rather than hallucinating when all strategies fail.
- Benchmarked standard vs. self-healing pipelines, characterizing when healing helps — it triggers on low-confidence queries at a∼3× token cost, improving hard cases while adding overhead where the baseline already succeeds.
Leaderboard Standings
Leaderboard Position Pending
Global test scores, peer standing percentiles, and algorithm leaderboard ranks are updated dynamically.
Assessment Highlights
Assessments Not Completed
Coding evaluations, system assessment results, and conceptual score badges will appear here after taking a test.
AI Collaboration Score
AI Collaboration Score Pending
Developer coding behavior, assistant cooperation, and AI pair-programming indicators are evaluated during live coding sessions.
Role Compatibility Profile
Role Compatibility Analysis Pending
Custom matching reports, candidate role compatibility percentiles, and core engineer strength profiles are processed once conceptual code screenings are complete.
Achievements
Achievements Not Earned
Special rewards, developer badges, system recognition certificates, and conceptual milestones will display here.
About Details
Professional Bio
Computer Science student with experience in full-stack development and AI/ML research. Skilled in building scalable web applications, vector search engines, and distributed systems. Passionate about optimizing performance and implementing robust AI pipelines.
B.Tech in Computer Science and Engineering (Artificial Intelligence)
Manipal Institute of Technology (2022 - 2026)
All India Senior School Certificate Examination (AISSCE)
The Millennium School ( - 2021)