Overview
Develop a Tax-Litigation Co-Pilot that transforms legal reasoning through AI. Work with Nobel Laureate Prof. Daron Acemoglu (MIT).
Role
Join The Justice Lab, Adalat AI's forward-thinking research unit developing a Tax-Litigation Co-Pilot that transforms legal reasoning through AI. Working under oversight from Nobel Laureate Prof. Daron Acemoglu (MIT), you'll turn cutting-edge ideas into production-ready systems that think like lawyers.
Responsibilities
Data & Evaluation
* Curate, label, and version corpora spanning four court tiers
* Create task sets for prediction, entailment, and explanation
* Build evaluation suites for factual consistency, citation faithfulness, and policy impact
Modeling & Experimentation
* Fine-tune and distill LLMs using RL, DPO, or SFT-style feedback
* Explore long-context and retrieval strategies (LoRA, RAG, chunking)
* Frame and formalise "legal reasoning" for outcome prediction
* Design experiments benchmarking LLMs on labeled tax-law and civil-procedure tasks
Legal-Reasoning Research
* Model precedential hierarchies and detect conflicts
* Generate citation-grounded chains of thought
* Prototype retrieval-augmented generation and evidence tracing systems
* Build causal inference pipelines
Collaboration & Documentation
* Present findings to Professors Acemoglu, Kang, and policy advisors
* Maintain reproducible logs and polished reports
* Co-author internal memos and external papers with academic partners
* Ship lightweight services or notebooks for engineering teams
Requirements
Essential
* 2+ years NLP/ML research experience (industry or graduate school)
* Fluency in PyTorch/JAX and modern LLM fine-tuning stacks
* Experience with large-corpus data wrangling and evaluation pipeline building
* Clear, concise technical writing and communication skills
Desirable
* Prior work on legal or policy datasets
* Publications at ACL, ICML, NeurIPS, or similar conferences
* Experience with causal-inference or decision-theoretic ML
* Familiarity with Indian tax or civil-procedure law
Benefits
* Work from home with flexible hours
* Unlimited PTO and generous vacation policy
* Autonomy and ownership over projects
* Learning & development resources
* Access to Harvard/MIT/Oxford academic networks
* Maternity and paternity leave
* Smart, humble, and friendly team environment
Apply
To apply, please send your resume and a cover letter with the subject line: "ML Research Engineer - Legal Reasoning".
#J-18808-Ljbffr