Chamel Nadir BouachaChamel Nadir Bouacha
AI researcher and software engineer
I work on retrieval and representation learning that preserve when things were true, especially in longitudinal healthcare data.
Research
Relevant is not always valid.
The Télécom SudParis internship introduced me to temporal RAG for healthcare. While surveying the field for my Master’s review thesis, I became interested in a problem: retrieved evidence can be semantically relevant yet wrong in time. Building the system for my State Engineer thesis made that problem concrete. TempMedBench grew from that work as a way to evaluate it.
TempMedBench
A synthetic patient-timeline retrieval benchmark and generation pipeline for temporally faithful healthcare retrieval and reasoning. Structured longitudinal patient timelines are built from synthetic patient records, then used to generate temporally dependent questions, gold evidence relevance judgements, and model-ready note chunks against which retrieval strategies are evaluated.
Temporal RAG for Healthcare
A retrieval-first architecture that combines semantic relevance with explicit temporal constraints over longitudinal healthcare information, addressing temporal granularity, event-time anchoring, provenance, and retrieval-time consistency.
Temporal annotation pipeline
A temporal annotation pipeline extended from temporal-expression detection and normalization to event extraction, event-to-time linking, and document-creation-time anchoring. The annotator combines rules, spaCy, and a small language model. Its annotation output contributed data used for the work that became WaTE.
Papers & theses
WaTE: Continuous Distributional Temporal Representations for Text Encoders
Accepted short paper, 38th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2026).
Building a Temporal Retrieval-Augmented Generation System for Healthcare Applications
ESI Algiers and Télécom SudParis, defended 22 June 2026.
Exploring Temporal Retrieval-Augmented Generation: Methods, Evaluation, and Open Challenges
ESI Algiers, defended 22 June 2026.
Engineering
I build the systems my research needs and apply the same engineering discipline at work.
Python/FastMCP server
Built and deployed a Python/FastMCP server converting existing repository business logic and skills into MCP tools, resources, and prompts for internal operations users.
IRCHAD — Indoor Navigation for the Visually Impaired
Indoor navigation system for visually impaired users, built as a fourth-year ESI team project across twelve repositories covering web, mobile, and backend services.
Team Lead and AI Developer for the team; built the architectural-floor-plan preprocessing pipeline and designed the evaluation protocol.
Evidence89% mAP for doors, walls, and windows
Automatic Code Optimization
Built a pipeline that converts PyTorch models into MLIR intermediate representations using the Linalg and TOSA dialects.
Trajectory
- Momentum Worldwide — Junior Full-Stack Developer
- Télécom SudParis — Research collaboration
- Télécom SudParis — Research Intern
- Maystro Delivery — Agentic AI Intern
- New York University Abu Dhabi (NYUAD) — Research Intern, Automatic Code Optimization
- École Nationale Supérieure d'Informatique (ESI) — State Engineer Degree and Academic Master
Teaching & recognition
Designed and delivered technical workshops for developer and student audiences: RAG in the era of LLMs, agentic AI, fundamentals of NLP, cloud and Azure fundamentals, and linear regression. Across these workshops, I have trained more than 100 people.
1st place: DataHack Datathon, 3rd Edition (Feb. 2026); Samsung Innovation Campus Capstone Project (Jan. 2026); HAICK26 Datathon (Dec. 2025).