Job Description

Marmon Technologies India Private Limited
As a part of the global industrial organization Marmon Holdings--which is backed by Berkshire Hathaway--you'll be doing things that matter, leading at every level, and winning a better way. We're committed to making a positive impact on the world, providing you with diverse learning and working opportunities, and fostering a culture where everyone's empowered to be their best.
We are seeking a professional with a bachelor's degree in B.Tech, B. E, B.Sc., BCom, BA, or LL.B. A post-graduate qualification in Analytics, Data, or Legal Operations is optional but not mandatory. The ideal candidate will have 4-7 years of experience in data and document analytics, with exposure to legal text formats being an added advantage. This role does not require drafting, negotiation, or fallback expertise. Candidates should possess hands-on experience in LLM-assisted classification or Python-based text parsing, focusing on operational execution rather than architecture. Additionally, prior experience working with global teams across multiple time zones is preferred.
Primary Function:
This role extracts and synthesizes clause -level redline patterns from Marmon templates and
historical negotiations. Deliverables center on structured intelligence --variant clusters,
exception frequency, edit rationales, and drift over time --to inform contract SMEs and future
playbook authors. The role does not draft fallback language or determine legal positions; it
provides the evidence and pattern clarity required for those decisions.
Secondary Function (as bandwidth allows):
Beyond contract -pattern analytics, this role may support adjacent legal data initiatives using the
same core skillset (text analytics, trend detection, pattern clustering). Examples include matter -
volume trends, help -desk inquiry grouping, metadata normalization, and other structured insight
work that benefits from pattern analysis, not legal decision-making.
This role does not draft playbooks, define fallback positions, or configure AI/CLM rule logic.
ESSENTIAL FUNCTIONS:
Redline Pattern Extraction & Variant Discovery

• Analyze historical tracked changes and markups at scale to identify recurring edit patterns,

counterparty tendencies, and structured clause -variation clusters in legal text.

• Group similar edits into normalized variant sets without determining fallback positions or legal acceptability.

Clause Change Clustering & Exception Mapping

• Classify edits by frequency, type (add / delete / narrow / broaden), and location in the document.
• Surface anomalies and outliers for legal SME review, not interpretation.

LLM -Assisted Summarization & Tagging

• Use AI tools to accelerate classification of edit intent and theme categorization while

maintaining analyst oversight and QC.

• Feed structured outputs to legal SMEs; no drafting or position -setting.

Insight Packaging for SMEs & Playbook Authors

• Produce clean variant summaries, drift reports, and exception trend snapshots to support

SME decision-making (not authoring).

• Deliver contract -type-specific insight packs (e.g., 200 -500 agreements) summarizing top

recurring edits, variant clusters, exception patterns, and drift.

• Present structured findings to SMEs who determine final positions.

Scalable Text Extraction & Data Normalization

• Use Python -based scripts and approved AI services to support extraction, normalization,

and clustering at scale (operate/iterate, not architect).

• Extract previous text vs revised language across batches of Word documents.
• Treat redlines as structured text elements, not legal judgments.

Secondary Analytics (As Bandwidth Allows)

• Support adjacent analytics initiatives (e.g., help-desk patterning, matter trend clustering, metadata normalization) using the same text -pattern skillset.
• Strictly analytics, no contract drafting, negotiation, or rule design.

Cross-Time -Zone Collaboration

• Provide clear async updates, backlog transparency, and pattern summaries to US-based legal stakeholders.

EDUCATION AND EXPERIENCE:

• Bachelor's degree (B.Tech / B. E / B.Sc. / BCom / BA or LL. B).
• Postgrad in Analytics, Data, Legal Ops optional, not required.
• 4-7 years in data/document analytics; exposure to legal text formats helpful.
• No drafting, negotiation, or fallback expertise required.
• Experience with LLM -assisted classification or Python-based text parsing preferred

(operate, not architect).

• Experience working with global teams across time zones preferred.

JOB PREREQUISITES:

• Ability to treat legal language as structured text (not interpretive meaning).
• Comfortable extracting & grouping tracked changes at scale.
• Accuracy mindset for clustering and outlier surfacing.
• Strong communication skills to present findings to SMEs for legal interpretation.
• Key guardrail: does not determine acceptability, fallbacks, or drafting.

Following receipt of a conditional offer of employment, candidates will be required to complete additional job-related screening processes as permitted or required by applicable law.

Skills Required