FE · 08 tools
Frontend
Interfaces & interaction
Next.js 15ReactJSTypeScriptJavaScriptTailwind CSSFramer Motionshadcn/uiHTML / CSS
available for opportunities · United Kingdom
hi there, i'm
JacinthMahanta.
>
I'm a London-based AI/ML Engineer, LLM/NLP Researcher, and Software Developer with an MSc in Artificial Intelligence from Lancaster University. My work spans applied AI, language models, agentic systems, backend APIs, and full-stack development. I recently completed an AI research placement with Lancashire Teaching Hospitals NHS Foundation Trust, where I worked on PEFT-based language models and agentic AI workflows for NHS pathway validation. Previously, I worked as a Junior API Developer with Lancaster University's ITPI team.
4+
Years Experience
30+
Projects Shipped
5+
Open Source
scroll
// 01. about me
The person behind the commits.
AI researcher and software developer focused on applied AI, language models and production software.
PROFILE_RECORD
Jacinth Mahanta
I'm Jacinth Mahanta, an AI/ML Researcher and Software Developer from India, currently based in London, UK.
I completed an MSc in Artificial Intelligence at Lancaster University in 2026, after earning a B.Tech in Electronics and Communication Engineering from NIT Silchar with a CGPA of 8.18/10.
I recently completed an AI research placement with Lancashire Teaching Hospitals NHS Foundation Trust, where I worked on Parameter-Efficient Fine-Tuning and agentic AI for NHS Referral to Treatment pathway validation. My work combined LoRA-adapted language models with clinical NLP, FastMCP, LangChain, and synthetic OMOP healthcare data.
I previously worked as a Junior API Developer with Lancaster University's ITPI team, developing REST APIs, backend integrations, and production-ready software features. My wider technical experience spans full-stack development, machine learning, healthcare NLP, cybersecurity, and UI/UX design.
Beyond development and research, I've served as a Technical Advisor & Outreach Head at GDGC NIT Silchar, mentored developers through open-source programmes, judged hackathons, and volunteered with the British Heart Foundation.
Current focus
What I am learning, building and looking for now.
01Education
MSc Artificial Intelligence · Lancaster University · 2026
02Researching
LLMs, PEFT, Healthcare NLP & Agentic AI
03Recently
AI Researcher with Lancashire Teaching Hospitals NHS Foundation Trust
04Building
AI/ML systems, APIs & full-stack applications
05Open to
Graduate AI/ML, Applied AI & Software Engineering roles
06UK roles
Skilled Worker visa sponsorship required
SYSTEM://PROFILE
Developer dashboard
ACADEMIC
0.00/10
Academic CGPA
CAPABILITY
0+
Core Domains
LANCASTER
MSc AI
Highest Degree
UK
London
Base
Current position
London, United Kingdom
Current track
Applied AI · LLMs · Software
Career trajectory
India → Lancaster → London · Applied AI
NIT Silchar
2021—25B.Tech · Electronics & Communication Engineering
India
Lancaster University
2025—26MSc · Artificial Intelligence
Lancaster, UK
NHS AI Research
2026PEFT · Healthcare NLP · Agentic AI
Lancashire Teaching Hospitals NHS FT
Applied AI / ML
NEXTGraduate AI/ML & Software Engineering
United Kingdom
CAPABILITY_MATRIX
What I bring to a team
Hover or focus a module to inspect the capability profile.
MODULE::AI
PyTorch, Hugging Face Transformers, PEFT/LoRA, LangChain, FastMCP, MLflow, NLP
// 02. tech stack
Tools of the trade.
A focused stack for AI research, production backend systems, modern interfaces and reliable delivery — not just a list of technologies.
stack topology
AI · Backend · Frontend · Delivery
core.runtime
AI + Software
research → systems → product
One stack across model development, backend services, interfaces and deployment — selected for building usable AI systems end to end.
04
modules
41
technologies
01
system
active.moduleAI
active capability
AI & Machine Learning
LLMs, NLP & applied ML
PythonPyTorchHugging Face TransformersPEFT / LoRALangChainFastMCPscikit-learn+6 more
BE · 10 tools
Backend & Databases
APIs, data & infrastructure
Node.jsExpress.jsREST APIsPostgreSQLMongoDBDynamoDBPrismaSupabaseAWSDocker
AI · 13 tools
AI & Machine Learning
LLMs, NLP & applied ML
PythonPyTorchHugging Face TransformersPEFT / LoRALangChainFastMCPscikit-learnMLflowNLPLLMsDuckDBIbisJupyter
DX · 10 tools
Design & Tools
Workflow, delivery & product
FigmaAdobe XDPrototypingDesign SystemsGitGitHubGitHub ActionsCI/CDVercelPostman
also working with
auxiliary toolkit
NextAuthClerkStripeResendInngestArcjetZodZustandReact Query
// 03. projects
Things I've built.
Selected work across clinical AI, LLM systems, machine learning, data engineering and full-stack product development.
✦featured projects
featured
94.7%
held-out accuracy
SYS://clinical
Clinical AI · LLM Systems
NHS RTT Pathway Validation with PEFT & Agentic AI
Built a clinical AI research pipeline for NHS Referral to Treatment pathway validation using synthetic OMOP CDM v5.4 data and outpatient letters. Integrated FastMCP data-access and specialty-guidance servers with a LangChain Deep Agent, and fine-tuned Qwen3.5-9B using LoRA and Hugging Face PEFT for three-class PIFU classification. The fine-tuned model achieved 94.7% accuracy and 0.947 macro F1 on a 150-case held-out evaluation set.
Qwen3.5-9BPEFT / LoRAHugging FaceLangChainFastMCPOMOP CDMMLflowHealthcare NLP
featured
30-day
readmission task
SYS://nlp
Clinical NLP · Transformers
Disease Progression Prediction with Clinical Transformers
A clinical NLP project predicting 30-day hospital readmission from MIMIC-IV EHR narratives, comparing TF-IDF baselines with fine-tuned transformers, and using attention analysis to identify clinically meaningful risk signals.
PythonNLPTransformersPubMedBERTBioBERTMIMIC-IVClinical AIMachine Learning
featured
Hybrid
recommendation stack
SYS://recsys
Big Data · Recommender Systems
Scalable E-Commerce Recommendation Pipeline
A big data analytics pipeline for Amazon Reviews 2023, benchmarking databases, creating embeddings, running similarity search and clustering, and generating personalized recommendations with collaborative, content-based, and hybrid models.
◇other noteworthy projects
archive
Deep Learning · Climate
Basel Climate DNN Analysis
A deep learning project for analyzing and predicting climate patterns using neural networks, with focus on data processing and visualization.
PythonTensorFlowPandasNumPyMatplotlib
archive
AI Safety · Governance
Societal AI Perspectives on AGI Evaluation
A research project analyzing AGI progress through benchmarks, impact measures, safety thresholds, and governance frameworks, addressing benchmark limits, data contamination, Goodharting, and advanced AI’s economic, ethical, and regulatory implications.
Artificial IntelligenceAGIAI GovernanceAI SafetyBenchmarksEthicsPolicy Research
archive
// 04. experience
Where I've worked.
A timeline spanning clinical AI research, backend and full-stack engineering, cybersecurity, technical leadership, mentorship, and community work.
08
roles
05
domains
04+
years
02
countries
Clinical AI Researchlatest
AI Researcher — PEFT for NHS RTT Pathway Validation
Lancashire Teaching Hospitals NHS Foundation TrustJun 2026 — Sep 2026
Lancaster, UK · Remote
Developed an agentic clinical AI pipeline for NHS Referral to Treatment pathway validation, integrating synthetic OMOP CDM v5.4 data with FastMCP, DuckDB, Ibis, and LangChain Deep Agents. Fine-tuned Qwen3.5-9B using LoRA and Hugging Face PEFT for outpatient-letter classification, achieving 94.7% accuracy and 0.947 macro F1 on a 150-case held-out evaluation set. Used MLflow for reproducible experiment tracking and evaluation, with emphasis on human review, clinical relevance, responsible AI, and error analysis.
PEFT / LoRAQwen3.5-9BHealthcare NLPAgentic AILangChain & FastMCPOMOP CDMRTT Pathway ValidationMLflow
Dec 2025 — Jul 2026
Lancaster, UK
Supported day-to-day retail operations, donation handling, and fundraising activities while contributing to the British Heart Foundation's community engagement and awareness initiatives.
Retail OperationsCustomer ServiceDonation HandlingFundraisingCommunity Engagement
Dec 2025 — Mar 2026
Lancaster, UK
Developed and maintained RESTful APIs supporting internal university services and student-facing digital platforms. Integrated backend services, optimised database queries and API performance, and contributed to testing, code review, and deployment of production-ready features.
REST APIsBackend DevelopmentAPI IntegrationPerformance OptimisationAPI Testing
Aug 2023 — May 2025
Silchar, India
Led technical outreach initiatives and mentored students across web development, UI/UX design, and cloud technologies. Organised workshops, bootcamps, hackathons, and open-source programmes while advising project teams on product design, development, and deployment workflows.
Technical LeadershipMentorshipUI/UX DesignDeveloper OutreachEvent ManagementFigma
// 05. certifications
Credentials I've earned.
Selected certifications and professional credentials across artificial intelligence, machine learning, software engineering and research.
17 credentials
Anthropic
Claude Academy: Model Context Protocol: Advanced topics
Issued October 2026
AI AgentsLarge Language Models (LLMs)
view credential
Google Analytics Certification
Issued October 2026
Data AnalysisStatistical Data AnalysisData Visualization
view credential05recommendations
Trusted by people who have worked beside me.
Feedback from collaborators and peers across software engineering, university projects, and AI research.
I worked with Jacinth on the SCC.450 group project, “Benchmarking Frameworks: Statistical and Machine Learning Approaches,” completed in collaboration with AMD. I was the project manager, and he was one of the people I could rely on most. Jacinth took full ownership of transferring our findings from a synthetic environment to real-world data. He handled data selection, preprocessing, and pipeline preparation. He delivered on time and with the research and implementation depth the work required. He then presented the findings to our team and later to the Programme Directors. He is an excellent communicator and a strong team player, always ready to help or advise colleagues. His contribution played a significant part in our project being selected for a celebratory presentation to the Programme Directors, the Module Leader, and the Head of the Data Science and AI Institute (DSAIL) at Lancaster University. I would recommend him highly for AI, Data Science, and Software Engineering roles.
Yulian Shkrum
Research Associate (Multi-Agent Autonomy) | MSc Artificial Intelligence LU | ex-President LUUASOC | Careers Manager USU · Lancaster University
Studied together at Lancaster University
// 06. contact
availableHave something worthbuilding together?
I'm open to AI/ML, LLM/NLP and software engineering opportunities, research collaborations, and technically interesting projects. Send the context and I'll get back to you.
message.transmission
secure contact channel